The role of spatial epidemiology to support public health policies : case studies applied to health promotion, noncommunicable and infectious diseases in the canton of Vaud, Switzerland

Towards Language Learning From Passive Exposures To In-Context Examples

Generalization and Personalization of Machine Learning for Multimodal Mobile Sensing in Everyday Life

Infectious disease spread in connected communities

Privacy-preserving contact tracing curbed COVID

M. Salathe 

Despite controversies over decentralized contact-tracing apps, the data now show that they saved thousands of lives during the pandemic. National and international authorities must heed the lessons.

Hits and Misses: Digital Contact Tracing in a Pandemic

M. Garry; R. Zajac; L. Hope; M. Salathe; L. Levine et al. 

Traditional contact tracing is one of the most powerful weapons people have in the battle against a pandemic, especially when vaccines do not yet exist or do not afford complete protection from infection. But the effectiveness of contact tracing hinges on its ability to find infected people quickly and obtain accurate information from them. Therefore, contact tracing inherits the challenges associated with the fallibilities of memory. Against this backdrop, digital contact tracing is the “dream scenario”-an unobtrusive, vigilant, and accurate recorder of danger that should outperform manual contact tracing on every dimension. There is reason to celebrate the success of digital contact tracing. Indeed, epidemiologists report that digital contact tracing probably reduced the incidence of COVID-19 cases by at least 25% in many countries, a feat that would have been hard to match with its manual counterpart. Yet there is also reason to speculate that digital contact tracing delivered on only a fraction of its potential because it almost completely ignored the relevant psychological science. We discuss the strengths and weaknesses of digital contact tracing, its hits and misses in the COVID-19 pandemic, and its need to be integrated with the science of human behavior.

COVID-Twitter-BERT: A natural language processing model to analyse COVID-19 content on Twitter

M. Muller; M. Salathe; P. E. Kummervold 

IntroductionThis study presents COVID-Twitter-BERT (CT-BERT), a transformer-based model that is pre-trained on a large corpus of COVID-19 related Twitter messages. CT-BERT is specifically designed to be used on COVID-19 content, particularly from social media, and can be utilized for various natural language processing tasks such as classification, question-answering, and chatbots. This paper aims to evaluate the performance of CT-BERT on different classification datasets and compare it with BERT-LARGE, its base model. MethodsThe study utilizes CT-BERT, which is pre-trained on a large corpus of COVID-19 related Twitter messages. The authors evaluated the performance of CT-BERT on five different classification datasets, including one in the target domain. The model’s performance is compared to its base model, BERT-LARGE, to measure the marginal improvement. The authors also provide detailed information on the training process and the technical specifications of the model. ResultsThe results indicate that CT-BERT outperforms BERT-LARGE with a marginal improvement of 10-30% on all five classification datasets. The largest improvements are observed in the target domain. The authors provide detailed performance metrics and discuss the significance of these results. DiscussionThe study demonstrates the potential of pre-trained transformer models, such as CT-BERT, for COVID-19 related natural language processing tasks. The results indicate that CT-BERT can improve the classification performance on COVID-19 related content, especially on social media. These findings have important implications for various applications, such as monitoring public sentiment and developing chatbots to provide COVID-19 related information. The study also highlights the importance of using domain-specific pre-trained models for specific natural language processing tasks. Overall, this work provides a valuable contribution to the development of COVID-19 related NLP models.

Dynamics of social media behavior before and after SARS-CoV-2 infection

F. Durazzi; F. Pichard; D. Remondini; M. Salathe 

IntroductionOnline social media have been both a field of research and a source of data for research since the beginning of the COVID-19 pandemic. In this study, we aimed to determine how and whether the content of tweets by Twitter users reporting SARS-CoV-2 infections changed over time. MethodsWe built a regular expression to detect users reporting being infected, and we applied several Natural Language Processing methods to assess the emotions, topics, and self-reports of symptoms present in the timelines of the users. ResultsTwelve thousand one hundred and twenty-one twitter users matched the regular expression and were considered in the study. We found that the proportions of health-related, symptom-containing, and emotionally non-neutral tweets increased after users had reported their SARS-CoV-2 infection on Twitter. Our results also show that the number of weeks accounting for the increased proportion of symptoms was consistent with the duration of the symptoms in clinically confirmed COVID-19 cases. Furthermore, we observed a high temporal correlation between self-reports of SARS-CoV-2 infection and officially reported cases of the disease in the largest English-speaking countries. DiscussionThis study confirms that automated methods can be used to find digital users publicly sharing information about their health status on social media, and that the associated data analysis may supplement clinical assessments made in the early phases of the spread of emerging diseases. Such automated methods may prove particularly useful for newly emerging health conditions that are not rapidly captured in the traditional health systems, such as the long term sequalae of SARS-CoV-2 infections.

Timeliness of online COVID-19 reports from official sources

L. Espinosa; O. Altunina; M. Salathe 

IntroductionMaking epidemiological indicators for COVID-19 publicly available through websites and social media can support public health experts in the near-real-time monitoring of the situation worldwide, and in the establishment of rapid response and public health measures to reduce the consequences of the pandemic. Little is known, however, about the timeliness of such sources. Here, we assess the timeliness of official public COVID-19 sources for the WHO regions of Europe and Africa. MethodsWe monitored official websites and social media accounts for updates and calculated the time difference between daily updates on COVID-19 cases. We covered a time period of 52 days and a geographic range of 62 countries, 28 from the WHO African region and 34 from the WHO European region. ResultsThe most prevalent categories were social media updates only (no website reporting) in the WHO African region (32.7% of the 1,092 entries), and updates in both social media and websites in the WHO European region (51.9% of the 884 entries for EU/EEA countries, and 73.3% of the 884 entries for non-EU/EEA countries), showing an overall clear tendency in using social media as an official source to report on COVID-19 indicators. We further show that the time difference for each source group and geographical region were statistically significant in all WHO regions, indicating a tendency to focus on one of the two sources instead of using both as complementary sources. DiscussionPublic health communication via social media platforms has numerous benefits, but it is worthwhile to do it in combination with other, more traditional means of communication, such as websites or offline communication.

Empathetic Conversational Agents for Distress Support

Deploying Decentralized, Privacy-Preserving Proximity Tracing

C. Troncoso; D. Bogdanov; E. Bugnion; S. Chatel; C. Cremers et al. 

The Food Recognition Benchmark: Using Deep Learning to Recognize Food in Images

S. P. Mohanty; G. Singhal; E. A. Scuccimarra; D. Kebaili; H. Heritier et al. 

The automatic recognition of food on images has numerous interesting applications, including nutritional tracking in medical cohorts. The problem has received significant research attention, but an ongoing public benchmark on non-biased (i.e., not scraped from web) data to develop open and reproducible algorithms has been missing. Here, we report on the setup of such a benchmark using publicly available food images sourced through the mobile MyFoodRepo app used in research cohorts. Through four rounds, the benchmark released the MyFoodRepo-273 dataset constituting 24,119 images and a total of 39,325 segmented polygons categorized in 273 different classes. Models were evaluated on private tests sets from the same platform with 5,000 images and 7,865 annotations in the final round. Top-performing models on the 273 food categories reached a mean average precision of 0.568 (round 4) and a mean average recall of 0.885 (round 3), and were deployed in production use of the MyFoodRepo app. We present experimental validation of round 4 results, and discuss implications of the benchmark setup designed to increase the size and diversity of the dataset for future rounds.

Associations Between Device-Measured Physical Activity and Glycemic Control and Variability Indices Under Free-Living Conditions

D. El Fatouhi; H. Heritier; C. Allemann; L. Malisoux; N. Laouali et al. 

Background: Disturbances of glycemic control and large glycemic variability have been associated with increased risk of type 2 diabetes in the general population as well as complications in people with diabetes. Long-term health benefits of physical activity are well documented but less is known about the timing of potential short-term effects on glycemic control and variability in free-living conditions.Materials and Methods: We analyzed data from 85 participants without diabetes from the Food & You digital cohort. During a 2-week follow-up, device-based daily step count was studied in relationship to glycemic control and variability indices using generalized estimating equations. Glycemic indices, evaluated using flash glucose monitoring devices (FreeStyle Libre), included minimum, maximum, mean, standard deviation, and coefficient of variation of daily glucose values, the glucose management indicator, and the approximate area under the sensor glucose curve.Results: We observed that every 1000 steps/day increase in daily step count was associated with a 0.3588 mg/dL (95% confidence interval [CI]: -0.6931 to -0.0245), a 0.0917 mg/dL (95% CI: -0.1793 to -0.0042), and a 0.0022% (95% CI: -0.0043 to -0.0001) decrease in the maximum glucose values, mean glucose, and in the glucose management indicator of the following day, respectively. We did not find any association between daily step count and glycemic indices from the same day.Conclusions: Increasing physical activity level was linked to blunted glycemic excursions during the next day. Because health-related benefits of physical activity can be long to observe, such short-term physiological benefits could serve as personalized feedback to motivate individuals to engage in healthy behaviors.

Visible-Light-Driven Water Oxidation on Self-Assembled Metal-Free Organic@Carbon Junctions at Neutral pH

A. J. Olaya; J. S. Riva; D. Baster; W. O. Silva; F. Pichard et al. 

Sustainable water oxidation requires low-cost, stable, and efficient redox couples, photosensitizers, and catalysts. Here, we introduce the in situ self-assembly of metal-atom-free organic-based semiconductive structures on the surface of carbon supports. The resulting TTF/TTF center dot+@carbon junction (TTF = tetrathiafulvalene) acts as an all-in-one highly stable redox-shuttle/photosensitizer/molecular-catalyst triad for the visible-light-driven water oxidation reaction (WOR) at neutral pH, eliminating the need for metallic or organometallic catalysts and sacrificial electron acceptors. A water/butyronitrile emulsion was used to physically separate the photoproducts of the WOR, H+ and TTF, allowing the extraction and subsequent reduction of protons in water, and the in situ electrochemical oxidation of TTF to TTF center dot+ on carbon in butyronitrile by constant anode potential electrolysis. During 100 h, no decomposition of TTF was observed and O-2 was from the emulsion while H-2 was in the This work new for a new generation of metal-atom-free, low-cost, redox-driven water-splitting strategies.

Clusters of science and health related Twitter users become more isolated during the COVID-19 pandemic

F. Durazzi; M. Muller; M. Salathe; D. Remondini 

COVID-19 represents the most severe global crisis to date whose public conversation can be studied in real time. To do so, we use a data set of over 350 million tweets and retweets posted by over 26 million English speaking Twitter users from January 13 to June 7, 2020. We characterize the retweet network to identify spontaneous clustering of users and the evolution of their interaction over time in relation to the pandemic’s emergence. We identify several stable clusters (super-communities), and are able to link them to international groups mainly involved in science and health topics, national elites, and political actors. The science- and health-related super-community received disproportionate attention early on during the pandemic, and was leading the discussion at the time. However, as the pandemic unfolded, the attention shifted towards both national elites and political actors, paralleled by the introduction of country-specific containment measures and the growing politicization of the debate. Scientific super-community remained present in the discussion, but experienced less reach and became more isolated within the network. Overall, the emerging network communities are characterized by an increased self-amplification and polarization. This makes it generally harder for information from international health organizations or scientific authorities to directly reach a broad audience through Twitter for prolonged time. These results may have implications for information dissemination along the unfolding of long-term events like epidemic diseases on a world-wide scale.

Digital proximity tracing on empirical contact networks for pandemic control

G. Cencetti; G. Santin; A. Longa; E. Pigani; A. Barrat et al. 

Digital contact tracing is a relevant tool to control infectious disease outbreaks, including the COVID-19 epidemic. Early work evaluating digital contact tracing omitted important features and heterogeneities of real-world contact patterns influencing contagion dynamics. We fill this gap with a modeling framework informed by empirical high-resolution contact data to analyze the impact of digital contact tracing in the COVID-19 pandemic. We investigate how well contact tracing apps, coupled with the quarantine of identified contacts, can mitigate the spread in real environments. We find that restrictive policies are more effective in containing the epidemic but come at the cost of unnecessary large-scale quarantines. Policy evaluation through their efficiency and cost results in optimized solutions which only consider contacts longer than 15-20 minutes and closer than 2-3 meters to be at risk. Our results show that isolation and tracing can help control re-emerging outbreaks when some conditions are met: (i) a reduction of the reproductive number through masks and physical distance; (ii) a low-delay isolation of infected individuals; (iii) a high compliance. Finally, we observe the inefficacy of a less privacy-preserving tracing involving second order contacts. Our results may inform digital contact tracing efforts currently being implemented across several countries worldwide. Digital contact tracing is increasingly considered as one of the tools to control infectious disease outbreaks, in particular the COVID-19 epidemic. Here, the authors present a modeling framework informed by empirical high-resolution contact data to analyze the impact of digital contact tracing apps.

Supervised Learning Computer Vision Benchmark for Snake Species Identification From Photographs: Implications for Herpetology and Global Health

A. M. Durso; G. K. Moorthy; S. P. Mohanty; I. Bolon; M. Salathe et al. 

We trained a computer vision algorithm to identify 45 species of snakes from photos and compared its performance to that of humans. Both human and algorithm performance is substantially better than randomly guessing (null probability of guessing correctly given 45 classes = 2.2%). Some species (e.g., Boa constrictor) are routinely identified with ease by both algorithm and humans, whereas other groups of species (e.g., uniform green snakes, blotched brown snakes) are routinely confused. A species complex with largely molecular species delimitation (North American ratsnakes) was the most challenging for computer vision. Humans had an edge at identifying images of poor quality or with visual artifacts. With future improvement, computer vision could play a larger role in snakebite epidemiology, particularly when combined with information about geographic location and input from human experts.

On the use of applied machine learning and digital infrastructure to leverage social media data in health and epidemiology

Learning Self-Exciting Temporal Point Processes Under Noisy Observations

Toward a Common Performance and Effectiveness Terminology for Digital Proximity Tracing Applications

W. Lueks; J. Benzler; D. Bogdanov; G. Kirchner; R. Lucas et al. 

CrowdNotifier: Decentralized Privacy-Preserving Presence Tracing

W. Lueks; S. Gürses; M. Veale; E. Bugnion; M. Salathé et al. 

There is growing evidence that SARS-CoV-2 can be transmitted beyond close proximity contacts, in particular in closed and crowded environments with insufficient ventilation. To help mitigation efforts, contact tracers need a way to notify those who were present in such environments at the same time as infected individuals. Neither traditional human-based contact tracing powered by handwritten or electronic lists, nor Bluetooth-enabled proximity tracing can handle this problem efficiently. In this paper, we propose CrowdNotifier, a protocol that can complement manual contact tracing by efficiently notifying visitors of venues and events with SARS-CoV-2-positive attendees. We prove that CrowdNotifier provides strong privacy and abuse resistance, and show that it can scale to handle notification at a national scale.

Modeling infectious disease dynamics towards informed public health interventions, with applications on COVID-19 and cholera

Early evidence of effectiveness of digital contact tracing for SARS-CoV-2 in Switzerland

M. Salathe; C. L. Althaus; N. Anderegg; D. Antonioli; T. Ballouz et al. 

In the wake of the pandemic of coronavirus disease 2019 (COVID-19), contact tracing has become a key element of strategies to control the spread of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2). Given the rapid and intense spread of SARS-CoV-2, digital contact tracing has emerged as a potential complementary tool to support containment and mitigation efforts. Early modelling studies highlighted the potential of digital contact tracing to break transmission chains, and Google and Apple subsequently developed the Exposure Notification (EN) framework, making it available to the vast majority of smartphones. A growing number of governments have launched or announced EN-based contact tracing apps, but their effectiveness remains unknown. Here, we report early findings of the digital contact tracing app deployment in Switzerland. We demonstrate proof-of-principle that digital contact tracing reaches exposed contacts, who then test positive for SARS-CoV-2. This indicates that digital contact tracing is an effective complementary tool for controlling the spread of SARS-CoV-2. Continued technical improvement and international compatibility can further increase the efficacy, particularly also across country borders.

A digital reconstruction of the 1630–1631 large plague outbreak in Venice

G. Lazzari; G. Colavizza; F. Bortoluzzi; D. Drago; A. Erboso et al. 

The plague, an infectious disease caused by the bacterium Yersinia pestis, is widely considered to be responsible for the most devastating and deadly pandemics in human history. Starting with the infamous Black Death, plague outbreaks are estimated to have killed around 100 million people over multiple centuries, with local mortality rates as high as 60%. However, detailed pictures of the disease dynamics of these outbreaks centuries ago remain scarce, mainly due to the lack of high-quality historical data in digital form. Here, we present an analysis of the 1630–1631 plague outbreak in the city of Venice, using newly collected daily death records. We identify the presence of a two-peak pattern, for which we present two possible explanations based on computational models of disease dynamics. Systematically digitized historical records like the ones presented here promise to enrich our understanding of historical phenomena of enduring importance. This work contributes to the recently renewed interdisciplinary foray into the epidemiological and societal impact of pre-modern epidemics.

Assessing Public Opinion on CRISPR-Cas9: Combining Crowdsourcing and Deep Learning

M. Muller; M. Schneider; M. Salathe; E. Vayena 

Conclusions: Overall, this type of analysis can provide valuable and complementary insights into ongoing public debates, extending the traditional empirical bioethics toolset.

Self-Supervised Prototypical Transfer Learning for Few-Shot Classification

C. Medina; A. Devos; M. Grossglauser 

Recent advances in transfer learning and few-shot learning largely rely on annotated data related to the goal task during (pre-)training. However, collecting sufficiently similar and annotated data is often infeasible. Building on advances in self-supervised and few-shot learning, we propose to learn a metric embedding that clusters unlabeled samples and their augmentations closely together. This pre-trained embedding serves as a starting point for classification with limited labeled goal task data by summarizing class clusters and fine-tuning. Experiments show that our approach significantly outperforms state-of the-art unsupervised meta-learning approaches, and is on par with supervised performance. In a cross-domain setting, our approach is competitive with its classical fully supervised counterpart.

A research agenda for digital proximity tracing apps

V. W. Viktor; B. Sebastian; B. Edouard; P. M. Alan; S. Marcel et al. 

Keep calm and carry on vaccinating: Is anti-vaccination sentiment contributing to declining vaccine coverage in England?

M. Edelstein; M. Muller; S. Ladhani; J. Yarwood; M. Salathe et al. 

Background: In England, coverage for childhood vaccines have decreased since 2012/13 in the context of an increasingly visible anti-vaccination discourse. We determined whether anti-vaccination sentiment is the likely cause of this decline in coverage. Methods: Descriptive study triangulating a range of data sources (vaccine coverage, cross-sectional survey of attitudes towards vaccination, UK-specific Twitter social media) and assessing them against the following Bradford Hill criteria: strength of association, consistency, specificity, temporality, biological gradient and coherence. Results: Strength of association: compared with well-documented vaccine scares, the decline in childhood vaccination seen since 2012/13 is 4-20 times smaller;consistency: while coverage for completed courses of the hexavalent and meningococcal vaccines decreased by 0.5-1.2 percentage points (pp) between 2017 and 2019, coverage for the first dose of these vaccines increased 0.5-0.7 pp; specificity: Since 2012-13, coverage decreased for some vaccines (hexavalent, MMR, HPV, shingles) and increased for others (MenACWY, Td/IPV, antenatal pertussis, influenza in 2 years of children), with no agespecific patterns. Temporality and biological gradient: the decline in vaccine coverage was preceded by an increase in vaccine confidence and a decrease in the proportion of parents encountering antivaccination materials. Coherence: attitudes towards vaccination expressed on Twitter in the UK became increasingly positive between 2017 and 2019 as vaccine coverage for childhood vaccines decreased. Conclusions: In England, trends in vaccine coverage between 2012/13 and 2018/19 were not homogenous and varied in magnitude and direction according to vaccine, dose and region. In addition, confidence in vaccines increased during the same period. These findings are not compatible with anti-vaccination sentiment causing a decline in vaccine coverage In England. (C) 2020 The Authors. Published by Elsevier Ltd.

Covid-19 contact tracing: efficacy and privacy

A. Collins 

This article assesses the risk-risk trade-off between privacy and efficacy that the use of contact tracing apps entails. It argues for the use of privacy-preserving apps, but highlights potential weaknesses and cautions against allowing digital tracing to detract from other pillars in the pandemic response.

COVID-19 epidemic in Switzerland: on the importance of testing, contact tracing and isolation

S. Marcel; A. L. Christian; N. Richard; S. Silvia; H. Emma et al. 

Switzerland is among the countries with the highest number of coronavirus disease-2019 (COVID-19) cases per capita in the world. There are likely many people with undetected SARS-CoV-2 infection because testing efforts are currently not detecting all infected people, including some with clinical disease compatible with COVID-19. Testing on its own will not stop the spread of SARS-CoV-2. Testing is part of a strategy. The World Health Organization recommends a combination of measures: rapid diagnosis and immediate isolation of cases, rigorous tracking and precautionary self-isolation of close contacts. In this article, we explain why the testing strategy in Switzerland should be strengthened urgently, as a core component of a combination approach to control COVID-19.

Contact networks across ages: computational analysis of epidemiological dynamics

On the Design of a Youth-Led, Issue-Based, Crowdsourced Global Monitoring Framework for the SDGs

S. P. Mohanty; R. Ramaswamy; A. K. Duraiappah 

In this paper, we propose a novel methodology and design to contribute towards the achievement of the 17 Sustainable Development Goals (SDGs) adopted by member states of the United Nations for a better and more sustainable future for all. We particularly focus on achieving SDG 4.7—using education to ensure all learners acquire the knowledge and skills needed to promote sustainable development. We describe the design of a crowdsourced approach to monitor issues at a local level, and then use the insights gained to indicate how learning can be achieved by the entire community. We begin by encouraging local communities to identify issues that they are concerned about, with an assumption that any issue identified will fall within the purview of the 17 SDGs. Each issue is then tagged with a plurality of actions taken to address it. Finally, we tag the positive or negative changes in the issue as perceived by members of the local community. This data is used to broadly indicate quantitative measures of community learning when solving a societal problem, in turn telling us how SDG 4.7 is being achieved. The paper describes the design of a unique, youth-led, technology-based, bottom-up approach, applicable to communities across the globe, which can potentially ensure transgressive learning through participation of and monitoring by the local community leading to sustainable development

Snakebite and snake identification: empowering neglected communities and health-care providers with AI

R. R. de Castaneda; A. M. Durso; N. Ray; J. L. Fernandez; D. J. Williams et al. 

Assessment of menstrual health status and evolution through mobile apps for fertility awareness

L. Symul; K. Wac; P. Hillard; M. Salathe 

For most women of reproductive age, assessing menstrual health and fertility typically involves regular visits to a gynecologist or another clinician. While these evaluations provide critical information on an individual’s reproductive health status, they typically rely on memory-based self-reports, and the results are rarely, if ever, assessed at the population level. In recent years, mobile apps for menstrual tracking have become very popular, allowing us to evaluate the reliability and tracking frequency of millions of self-observations, thereby providing an unparalleled view, both in detail and scale, on menstrual health and its evolution for large populations. In particular, the primary aim of this study was to describe the tracking behavior of the app users and their overall observation patterns in an effort to understand if they were consistent with previous small-scale medical studies. The secondary aim was to investigate whether their precision allowed the detection and estimation of ovulation timing, which is critical for reproductive and menstrual health. Retrospective self-observation data were acquired from two mobile apps dedicated to the application of the sympto-thermal fertility awareness method, resulting in a dataset of more than 30 million days of observations from over 2.7 million cycles for two hundred thousand users. The analysis of the data showed that up to 40% of the cycles in which users were seeking pregnancy had recordings every single day. With a modeling approach using Hidden Markov Models to describe the collected data and estimate ovulation timing, it was found that follicular phases average duration and range were larger than previously reported, with only 24% of ovulations occurring at cycle days 14 to 15, while the luteal phase duration and range were in line with previous reports, although short luteal phases (10 days or less) were more frequently observed (in up to 20% of cycles). The digital epidemiology approach presented here can help to lead to a better understanding of menstrual health and its connection to women’s health overall, which has historically been severely understudied.

WHO and ITU establish benchmarking process for artificial intelligence in health

T. Wiegand; R. Krishnamurthy; M. Kuglitsch; N. Lee; S. Pujari et al. 

Crowdbreaks: Tracking Health Trends Using Public Social Media Data and Crowdsourcing

M. M. Mueller; M. Salathe 

In the past decade, tracking health trends using social media data has shown great promise, due to a powerful combination of massive adoption of social media around the world, and increasingly potent hardware and software that enables us to work with these new big data streams. At the same time, many challenging problems have been identified. First, there is often a mismatch between how rapidly online data can change, and how rapidly algorithms are updated, which means that there is limited reusability for algorithms trained on past data as their performance decreases over time. Second, much of the work is focusing on specific issues during a specific past period in time, even though public health institutions would need flexible tools to assess multiple evolving situations in real time. Third, most tools providing such capabilities are proprietary systems with little algorithmic or data transparency, and thus little buy-in from the global public health and research community. Here, we introduce Crowdbreaks, an open platform which allows tracking of health trends by making use of continuous crowdsourced labeling of public social media content. The system is built in a way which automatizes the typical workflow from data collection, filtering, labeling and training of machine learning classifiers and therefore can greatly accelerate the research process in the public health domain. This work describes the technical aspects of the platform, thereby covering the functionalities at its current state and exploring its future use cases and extensions.

Wet Markets and Food Safety: TripAdvisor for Improved Global Digital Surveillance

N. E. Kogan; I. Bolon; N. Ray; G. Alcoba; J. L. Fernandez-Marquez et al. 

Background: Wet markets are markets selling fresh meat and produce. Wet markets are critical for food security and sustainable development in their respective regions. Due to their cultural significance, they attract numerous visitors and consequently generate tourist-geared information on the Web (ie, on social networks such as TripAdvisor). These data can be used to create a novel, international wet market inventory to support epidemiological surveillance and control in such settings, which are often associated with negative health outcomes. Objective: Using social network data, we aimed to assess the level of wet markets’ touristic importance on the Web, produce the first distribution map of wet markets of touristic interest, and identify common diseases facing visitors in these settings. Methods: A Google search was performed on 31 food market-related keywords, with the first 150 results for each keyword evaluated based on their relevance to tourism. Of all these queries, wet market had the highest number of tourism-related Google Search results; among these, TripAdvisor was the most frequently-occurring travel information aggregator, prompting its selection as the data source for this study. A Web scraping tool (ParseHub) was used to extract wet market names, locations, and reviews from TripAdvisor. The latter were searched for disease-related content, which enabled assignment of GeoSentinel diagnosis codes to each. This syndromic categorization was overlaid onto a mapping of wet market locations. Regional prevalence of the most commonly occurring symptom group – food poisoning – was then determined (ie, by dividing the number of wet markets per continent with more than or equal to 1 review containing this syndrome by the total number of wet markets on that continent with syndromic information). Results: Of the 1090 hits on TripAdvisor for wet market, 36.06% (393/1090) conformed to the query’s definition; wet markets were heterogeneously distributed: Asia concentrated 62.6% (246/393) of them, Europe 19.3% (76/393), North America 7.9% (31/393), Oceania 5.1% (20/393), Africa 3.1% (12/393), and South America 2.0% (8/393). Syndromic information was available for 14.5% (57/393) of wet markets. The most frequently occurring syndrome among visitors to these wet markets was food poisoning, accounting for 54% (51/95) of diagnoses. Cases of this syndrome were identified in 56% (22/39) of wet markets with syndromic information in Asia, 71% (5/7) in Europe, and 71% (5/7) in North America. All wet markets in South America and Oceania reported food poisoning cases, but the number of reviews with syndromic information was very limited in these regions (n=2). Conclusions: The map produced illustrates the potential role of touristically relevant social network data to support global epidemiological surveillance. This includes the possibility to approximate the global distribution of wet markets and to identify diseases (ie, food poisoning) that are most prevalent in such settings.

Chemokine profiling in serum from patients with ovarian cancer reveals candidate biomarkers for recurrence and immune infiltration

A. Mlynska; G. Salciuniene; K. Zilionyte; S. Garberyte; M. Strioga et al. 

The management of advanced ovarian cancer is challenging due to the high frequency of recurrence, often associated with the development of resistance to platinum-based chemotherapy. Molecular analyses revealed the complexity of ovarian cancer with particular emphasis on the immune system, which may contribute to disease progression and response to treatment. Cytokines and chemokines mediate the cross-talk between cancer and immune cells, and therefore, present as potential biomarkers, reflecting the tumor microenvironment. A panel of circulating C-C motif chemokine ligand (CCL) and C-X-C motif chemokine ligand (CXCL) chemokines were examined in the serum of 40 high-grade patients with ovarian cancer prior to primary surgery. The level of immune infiltration in tumors was also analyzed. The preoperative levels of chemokines differ between patients. Elevated levels of circulating CXCL4 + CCL20 + CXCL1 combination can discriminate patients with shorter recurrence-free survival and overall survival. The presence of tumor-infiltrating T lymphocytes was detected in half of the patients. The mRNA expression analysis suggests the presence of antitumoral and immunosuppressive elements in the tumor microenvironment. The combination of circulating CXCL9 + CXCL10 can distinguish immune-infiltrated tumors that will lead to shorter recurrence-free survival. The results suggest that preoperative profiling of circulating chemokines in patients with ovarian cancer may provide valuable information regarding tumor recurrence and immune infiltration. The findings demonstrate that combinations have better prognostic utility than single chemokines, and may serve as patient stratification tools.

Assessing the Dynamics and Control of Droplet- and Aerosol-Transmitted Influenza Using an Indoor Positioning System

T. Smieszek; G. Lazzari; M. Salathé 

There is increasing evidence that aerosol transmission is a major contributor to the spread of influenza. Despite this, virtually all studies assessing the dynamics and control of influenza assume that it is transmitted solely through direct contact and large droplets, requiring close physical proximity. Here, we use wireless sensors to measure simultaneously both the location and close proximity contacts in the population of a US high school. This dataset, highly resolved in space and time, allows us to model both droplet and aerosol transmission either in isolation or in combination. In particular, it allows us to computationally quantify the potential effectiveness of overlooked mitigation strategies such as improved ventilation that are available in the case of aerosol transmission. Our model suggests that recommendation-abiding ventilation could be as effective in mitigating outbreaks as vaccinating approximately half of the population. In simulations using empirical transmission levels observed in households, we find that bringing ventilation to recommended levels had the same mitigating effect as a vaccination coverage of 50% to 60%. Ventilation is an easy-to-implement strategy that has the potential to support vaccination efforts for effective control of influenza spread.

An Interactive Gameplay to Crowdsource Multiple Sequence Alignment of Genome Sequences: Genenigma

D. A. Meedeniya; S. A. P. A. Rukshan; A. Welivita 

Comparative genomics is a field of research that compares genomes of different organisms to identify common patterns. It is a powerful method used to identify the genetic diseases that cause mutations. Multiple Sequence Alignment (MSA) is an intermediate step in comparative genomics analysis that aligns three or more biological sequences of similar length. MSA is an NP-hard problem for which no efficient algorithm exists to perform this in a reasonable amount of time. However, humans across evolution have developed special intuition to identify visual patterns in short periods of time. Hence, a citizen science approach can be devised to solve the MSA problem by transforming it into a human computing game on creating visually similar patterns. In this paper, we introduce the mobile game “Genenigma”, which harnesses the human computing capability to align multiple sequences of genomes and use the results to help geneticists to understand the genetic code. The usability and performance scores of “Genenigma” predicts a larger user base than existing mobile games built for this purpose.

Translating Science Into Business Innovation: The Case of Open Food and Nutrition Data Hackathons

C. Tucci; G. Viscusi; H. C. Gautschi 

In this article, we explore the use of hackathons and open data in corporations’ open innovation portfolios, addressing a new way for companies to tap into the creativity and innovation of early-stage startup culture, in this case applied to the food and nutrition sector. We study the first Open Food Data Hackdays, held on 10–11 February 2017 in Lausanne and Zurich. The aim of the overall project that the Hackdays event was part of was to use open food and nutrition data as a driver for business innovation. We see hackathons as a new tool in the innovation manager’s toolkit, a kind of live crowdsourcing exercise that goes beyond traditional ideation and develops a variety of prototypes and new ideas for business innovation. Companies then have the option of working with entrepreneurs and taking some the ideas forward.

Digital epidemiology: what is it, and where is it going?

M. Salathe 

Digital Epidemiology is a new field that has been growing rapidly in the past few years, fueled by the increasing availability of data and computing power, as well as by breakthroughs in data analytics methods. In this short piece, I provide an outlook of where I see the field heading, and offer a broad and a narrow definition of the term.

Augmenting Research, Education, and Outreach with Client-Side Web Programming

L. Abriata; J. Rodrigues; M. Salathe; L. Patiny 

FoodRepo: An Open Food Repository of Barcoded Food Products

G. Lazzari; Y. Jaquet; D. Kebaili; L. Symul; M. Salathe 

Localizing the Source of an Epidemic Using Few Observations

Likelihood-free Inference of Population Genetic Parameters from Time-Sampled Genetic Data

A field-based modelling framework of the ecohydrology of schistosomiasis

An ecological and digital epidemiology analysis on the role of human behavior on the 2014 Chikungunya outbreak in Martinique

B. Roche; B. Gaillard; L. Leger; R. Pelagie-Moutenda; T. Sochacki et al. 

Understanding the spatio-temporal dynamics of endemic infections is of critical importance for a deeper understanding of pathogen transmission, and for the design of more efficient public health strategies. However, very few studies in this domain have focused on emerging infections, generating a gap of knowledge that hampers epidemiological response planning. Here, we analyze the case of a Chikungunya outbreak that occurred in Martinique in 2014. Using time series estimates from a network of sentinel practitioners covering the entire island, we first analyze the spatio-temporal dynamics and show that the largest city has served as the epicenter of this epidemic. We further show that the epidemic spread from there through two different propagation waves moving northwards and southwards, probably by individuals moving along the road network. We then develop a mathematical model to explore the drivers of the temporal dynamics of this mosquito-borne virus. Finally, we show that human behavior, inferred by a textual analysis of messages published on the social network Twitter, is required to explain the epidemiological dynamics over time. Overall, our results suggest that human behavior has been a key component of the outbreak propagation, and we argue that such results can lead to more efficient public health strategies specifically targeting the propagation process.

Critical dynamics in population vaccinating behavior

A. D. Pananos; T. M. Bury; C. Wang; J. Schonfeld; S. P. Mohanty et al. 

Vaccine refusal can lead to renewed outbreaks of previously eliminated diseases and even delay global eradication. Vaccinating decisions exemplify a complex, coupled system where vaccinating behavior and disease dynamics influence one another. Such systems often exhibit critical phenomena-special dynamics close to a tipping point leading to a new dynamical regime. For instance, critical slowing down (declining rate of recovery from small perturbations) may emerge as a tipping point is approached. Here, we collected and geocoded tweets about measles-mumps-rubella vaccine and classified their sentiment using machine-learning algorithms. We also extracted data on measles-related Google searches. We find critical slowing down in the data at the level of California and the United States in the years before and after the 2014-2015 Disneyland, California measles outbreak. Critical slowing down starts growing appreciably several years before the Disneyland outbreak as vaccine uptake declines and the population approaches the tipping point. However, due to the adaptive nature of coupled behavior-disease systems, the population responds to the outbreak by moving away from the tipping point, causing “critical speeding up” whereby resilience to perturbations increases. A mathematical model of measles transmission and vaccine sentiment predicts the same qualitative patterns in the neighborhood of a tipping point to greatly reduced vaccine uptake and large epidemics. These results support the hypothesis that population vaccinating behavior near the disease elimination threshold is a critical phenomenon. Developing new analytical tools to detect these patterns in digital social data might help us identify populations at heightened risk of widespread vaccine refusal.

Precision global health in the digital age

A. Flahault; A. Geissbuhler; I. Guessous; P. J. Guerin; I. Bolon et al. 

Precision global health is an approach similar to precision medicine, which facilitates, through innovation and technology, better targeting of public health interventions on a global scale, for the purpose of maximising their effectiveness and relevance. Illustrative examples include: the use of remote sensing data to fight vector-borne diseases; large databases of genomic sequences of foodborne pathogens helping to identify origins of outbreaks; social networks and internet search engines for tracking communicable diseases; cell phone data in humanitarian actions; drones to deliver healthcare services in remote and secluded areas. Open science and data sharing platforms are proposed for fostering international research programmes under fair, ethical and respectful conditions. Innovative education, such as massive open online courses or serious games, can promote wider access to training in public health and improving health literacy. The world is moving towards learning healthcare systems. Professionals are equipped with data collection and decision support devices. They share information, which are complemented by external sources, and analysed in real time using machine learning techniques. They allow for the early detection of anomalies, and eventually guide appropriate public health interventions. This article shows how information- driven approaches, enabled by digital technologies, can help improving global health with greater equity.

Spatially explicit modeling of cholera epidemics

Digital Pharmacovigilance and Disease Surveillance: Combining Traditional and Big-Data Systems for Better Public Health

M. Salathe 

The digital revolution has contributed to very large data sets (ie, big data) relevant for public health. The two major data sources are electronic health records from traditional health systems and patient-generated data. As the two data sources have complementary strengths-high veracity in the data from traditional sources and high velocity and variety in patient-generated data-they can be combined to build more-robust public health systems. However, they also have unique challenges. Patient-generated data in particular are often completely unstructured and highly context dependent, posing essentially a machine-learning challenge. Some recent examples from infectious disease surveillance and adverse drug event monitoring demonstrate that the technical challenges can be solved. Despite these advances, the problem of verification remains, and unless traditional and digital epidemiologic approaches are combined, these data sources will be constrained by their intrinsic limits.

Infectious Disease Containment Based on a Wireless Sensor System

X. Sun; Z. Lu; X. Zhang; M. Salathe; G. Cao 

Infectious diseases pose a serious threat to public health due to its high infectivity and potentially high mortality. One of the most effective ways to protect people from being infected by these diseases is through vaccination. However, due to various resource constraints, vaccinating all the people in a community is not practical. Therefore, targeted vaccination, which vaccinates a small group of people, is an alternative approach to contain infectious diseases. Since many infectious diseases spread among people by droplet transmission within a certain range, we deploy a wireless sensor system in a high school to collect contacts happened within the disease transmission distance. Based on the collected traces, a graph is constructed to model the disease propagation, and a new metric (called connectivity centrality) is presented to find the important nodes in the constructed graph for disease containment. Connectivity centrality considers both a node’s local and global effect to measure its importance in disease propagation. Centrality based algorithms are presented and further enhanced by exploiting the information of the known infected nodes, which can be detected during targeted vaccination. Simulation results show that our algorithms can effectively contain infectious diseases and outperform other schemes under various conditions.

Using Deep Learning for Image-Based Plant Disease Detection

S. P. Mohanty; D. P. Hughes; M. Salathe 

Crop diseases are a major threat to food security, but their rapid identification remains difficult in many parts of the world due to the lack of the necessary infrastructure. The combination of increasing global smartphone penetration and recent advances in computer vision made possible by deep learning has paved the way for smartphone-assisted disease diagnosis. Using a public dataset of 54,306 images of diseased and healthy plant leaves collected under controlled conditions, we train a deep convolutional neural network to identify 14 crop species and 26 diseases (or absence thereof). The trained model achieves an accuracy of 99.35% on a held-out test set, demonstrating the feasibility of this approach. Overall, the approach of training deep learning models on increasingly large and publicly available image datasets presents a clear path toward smartphone-assisted crop disease diagnosis on a massive global scale.

Statistical physics of vaccination

Z. Wang; C. T. Bauch; S. Bhattacharyya; A. D’Onofrio; P. Manfredi et al. 

Historically, infectious diseases caused considerable damage to human societies, and they continue to do so today. To help reduce their impact, mathematical models of disease transmission have been studied to help understand disease dynamics and inform prevention strategies. Vaccination one of the most important preventive measures of modern times is of great interest both theoretically and empirically. And in contrast to traditional approaches, recent research increasingly explores the pivotal implications of individual behavior and heterogeneous contact patterns in populations. Our report reviews the developmental arc of theoretical epidemiology with emphasis on vaccination, as it led from classical models assuming homogeneously mixing (mean-field) populations and ignoring human behavior, to recent models that account for behavioral feedback and/or population spatial/social structure. Many of the methods used originated in statistical physics, such as lattice and network models, and their associated analytical frameworks. Similarly, the feedback loop between vaccinating behavior and disease propagation forms a coupled nonlinear system with analogs in physics. We also review the new paradigm of digital epidemiology, wherein sources of digital data such as online social media are mined for high-resolution information on epidemiologically relevant individual behavior. Armed with the tools and concepts of statistical physics, and further assisted by new sources of digital data, models that capture nonlinear interactions between behavior and disease dynamics offer a novel way of modeling real-world phenomena, and can help improve health outcomes. We conclude the review by discussing open problems in the field and promising directions for future research. (C) 2016 Elsevier B.V. All rights reserved.

Identifying Adverse Effects of HIV Drug Treatment and Associated Sentiments Using Twitter

C. Adrover; T. Bodnar; Z. Huang; A. Telenti; M. Salathé 

Background: Social media platforms are increasingly seen as a source of data on a wide range of health issues. Twitter is of particular interest for public health surveillance because of its public nature. However, the very public nature of social media platforms such as Twitter may act as a barrier to public health surveillance, as people may be reluctant to publicly disclose information about their health. This is of particular concern in the context of diseases that are associated with a certain degree of stigma, such as HIV/AIDS. Objective: The objective of the study is to assess whether adverse effects of HIV drug treatment and associated sentiments can be determined using publicly available data from social media. Methods: We describe a combined approach of machine learning and crowdsourced human assessment to identify adverse effects of HIV drug treatment solely on individual reports posted publicly on Twitter. Starting from a large dataset of 40 million tweets collected over three years, we identify a very small subset (1642; 0.004%) of individual reports describing personal experiences with HIV drug treatment. Results: Despite the small size of the extracted final dataset, the summary representation of adverse effects attributed to specific drugs, or drug combinations, accurately captures well-recognized toxicities. In addition, the data allowed us to discriminate across specific drug compounds, to identify preferred drugs over time, and to capture novel events such as the availability of preexposure prophylaxis. Conclusions: The effect of limited data sharing due to the public nature of the data can be partially offset by the large number of people sharing data in the first place, an observation that may play a key role in digital epidemiology in general.

Ethical challenges of big data in public health

E. Vayena; M. Salathé; L. C. Madoff; J. S. Brownstein 

Modeling Individual-Level Infection Dynamics Using Social Network Information

S. Tuarob; C. Tucker; M. Salathé; N. Ram 

Epidemic monitoring systems engaged in accurate discovery of infected individuals enable better understanding of the dynamics of epidemics and thus may promote effective disease mitigation or prevention. Currently, infection discovery systems require either physical participation of potential patients or provision of information from hospitals and health-care services. While social media has emerged as an increasingly important knowledge source that reflects multiple real world events, there is only a small literature examining how social media information can be incorporated into computational epidemic models. In this paper, we demonstrate how social media information can be incorporated into and improve upon traditional techniques used to model the dynamics of infectious diseases. Using flu infection histories and social network data collected from 264 students in a college community, we identify social network signals that can aid identification of infected individuals. Extending the traditional SIRS model, we introduce and illustrate the efficacy of an Online-Interaction-Aware Susceptible-Infected-Recovered-Susceptible (OIA-SIRS) model based on four social network signals for modeling infection dynamics. Empirical evaluations of our case study, flu infection within a college community, reveal that the OIA-SIRS model is more accurate than the traditional model, and also closely tracks the real-world infection rates as reported by CDC ILINet and Google Flu Trend

Targeted vaccination based on a wireless sensor system

S. Xiao; L. Zongqing; Z. Xiaomei; M. Salathé; C. Guohong 

Vaccination is one of the most effective ways to protect people from being infected by infectious disease. However, it is often impractical to vaccinate all people in a community due to various resource constraints. Therefore, targeted vaccination, which vaccinates a small group of people, is an alternative approach to contain infectious disease spread. To achieve better performance in targeted vaccination, we collect student contact traces in a high school based on wireless sensors carried by students. With our wireless sensor system, we can record student contacts within the disease propagation distance, and then construct a disease propagation graph to model the infectious disease propagation. Based on this graph, we propose a metric called connectivity centrality to measure a node’s importance during disease propagation and design centrality based algorithms for targeted vaccination. The proposed algorithms are evaluated and compared with other schemes based on our collected traces. Trace driven simulation results show that our algorithms can help to effectively contain infectious disease

Measles Vaccination Coverage and Cases among Vaccinated Persons

C. L. Althaus; M. Salathé 

How should social mixing be measured: comparing web-based survey and sensor-based methods

T. Smieszek; V. C. Barclay; I. Seeni; J. J. Rainey; H. Gao et al. 

Contact surveys and diaries have conventionally been used to measure contact networks in different settings for elucidating infectious disease transmission dynamics of respiratory infections. More recently, technological advances have permitted the use of wireless sensor devices, which can be worn by individuals interacting in a particular social context to record high resolution mixing patterns. To date, a direct comparison of these two different methods for collecting contact data has not been performed.

An ensemble heterogeneous classification methodology for discovering health-related knowledge in social media messages

S. Tuarob; C. S. Tucker; M. Salathe; N. Ram 

The role of social media as a source of timely and massive information has become more apparent since the era of Web 2.0.Multiple studies illustrated the use of information in social media to discover biomedical and health-related knowledge.Most methods proposed in the literature employ traditional document classification techniques that represent a document as a bag of words.These techniques work well when documents are rich in text and conform to standard English; however, they are not optimal for social media data where sparsity and noise are norms.This paper aims to address the limitations posed by the traditional bag-of-word based methods and propose to use heterogeneous features in combination with ensemble machine learning techniques to discover health-related information, which could prove to be useful to multiple biomedical applications, especially those needing to discover health-related knowledge in large scale social media data.Furthermore, the proposed methodology could be generalized to discover different types of information in various kinds of textual data.

Targeting HIV-related Medication Side Effects and Sentiment Using Twitter Data

C. Adrover; T. Bodnar; M. Salathé 

We present a descriptive analysis of Twitter data. Our study focuses on extracting the main side effects associated with HIV treatments. The crux of our work was the identification of personal tweets referring to HIV. We summarize our results in an infographic aimed at the general public. In addition, we present a measure of user sentiment based on hand-rated tweets

Two Sides of a Coin: Separating Personal Communication and Public Dissemination Accounts in Twitter

P. Yin; N. Ram; W-C. Lee; C. Tucker; S. Khandelwal et al. 

There are millions of accounts in Twitter. In this paper, we categorize twitter accounts into two types, namely Personal Communication Account (PCA) and Public Dissemination Account (PDA). PCAs are accounts operated by individuals and are used to express that individual’s thoughts and feelings. PDAs, on the other hand, refer to accounts owned by non-individuals such as companies, governments, etc. Generally, Tweets in PDA (i) disseminate a specific type of information (e.g., job openings, shopping deals, car accidents) rather than sharing an individual’s personal life; and (ii) may be produced by non-human entities (e.g., bots). We aim to develop techniques for identifying PDAs so as to (i) facilitate social scientists to reduce “noise” in their study of human behaviors, and (ii) to index them for potential recommendation to users looking for specific types of information. Through analysis, we find these two types of accounts follow different temporal, spatial and textual patterns. Accordingly we develop probabilistic models based on these features to identify PDAs. We also conduct a series of experiments to evaluate those algorithms for cleaning the Twitter data stream

Positive network assortativity of influenza vaccination at a high school: implications for outbreak risk and herd immunity

V. C. Barclay; T. Smieszek; J. He; G. Cao; J. J. Rainey et al. 

Schools are known to play a significant role in the spread of influenza. High vaccination coverage can reduce infectious disease spread within schools and the wider community through vaccine-induced immunity in vaccinated individuals and through the indirect effects afforded by herd immunity. In general, herd immunity is greatest when vaccination coverage is highest, but clusters of unvaccinated individuals can reduce herd immunity. Here, we empirically assess the extent of such clustering by measuring whether vaccinated individuals are randomly distributed or demonstrate positive assortativity across a United States high school contact network. Using computational models based on these empirical measurements, we further assess the impact of assortativity on influenza disease dynamics. We found that the contact network was positively assortative with respect to influenza vaccination: unvaccinated individuals tended to be in contact more often with other unvaccinated individuals than with vaccinated individuals, and these effects were most pronounced when we analyzed contact data collected over multiple days. Of note, unvaccinated males contributed substantially more than unvaccinated females towards the measured positive vaccination assortativity. Influenza simulation models using a positively assortative network resulted in larger average outbreak size, and outbreaks were more likely, compared to an otherwise identical network where vaccinated individuals were not clustered. These findings highlight the importance of understanding and addressing heterogeneities in seasonal influenza vaccine uptake for prevention of large, protracted school-based outbreaks of influenza, in addition to continued efforts to increase overall vaccine coverage.

On the ground validation of online diagnosis with Twitter and medical records

T. Bodnar; V. Barclay; N. Ram; C. Tucker; M. Salathé 

Social media has been considered as a data source for tracking disease. However, most analyses are based on models that prioritize strong correlation with population-level disease rates over determining whether or not specific individual users are actually sick. Taking a different approach, we develop a novel system for social-media based disease detection at the individual level using a sample of professionally diagnosed individuals. Specifically, we develop a system for making an accurate influenza diagnosis based on an individual’s publicly available Twitter data. We find that about half (17/35 = 48.57%) of the users in our sample that were sick explicitly discuss their disease on Twitter. By developing a meta classifier that combines text analysis, anomaly detection, and social network analysis, we are able to diagnose an individual with greater than 99% accuracy even if she does not discuss her health

Synthesizing Social Proximity Networks by Combining Subjective Surveys with Digital Traces

H. Xia; J. Chen; M. Marathe; H. Mortveit; M. Salathé 

Synthetic social contact networks play a central role in the study of epidemics and methods to control them. In this paper we propose a new methodology that combines subjective surveys and data obtained using digital devices to synthesize detailed social networks for high schools in the United States. The two data sources are diverse and have their relative merits. The proposed methodology yields high quality dynamic social proximity networks. We evaluate our methodology by carrying out a detailed structural analysis of the resulting networks. Epidemic simulations and intervention analysis using these networks provide further insights into the role of network structure on epidemics. Our results indicate that the in-class networks have a highly clustered structure with contact duration following a heavy tail distribution. SEIR-based epidemic simulations demonstrate that we may use existing theoretic graph models to fit digital trace in-class networks, but only after critical structure metrics including degree and edge weight are tuned to the real data. For practical use, the detailed model for in-class contacts using digital trace data therefore seems to add important and valuable structure needed when developing public health policies. Our methodology is quite general and can be combined with subjective assessments such as surveys and other available information. The technique is also applicable to other micro-networks such as conferences with multiple sessions, and office campuses. It is efficient and applicable in settings where data is hard or relatively expensive to obtain

A low-cost method to assess the epidemiological importance of individuals in controlling infectious disease outbreaks

T. Smieszek; M. Salathé 

Infectious disease outbreaks in communities can be controlled by early detection and effective prevention measures. Assessing the relative importance of each individual community member with respect to these two processes requires detailed knowledge about the underlying social contact network on which the disease can spread. However, mapping social contact networks is typically too resource-intensive to be a practical possibility for most communities and institutions.

Discovering health-related knowledge in social media using ensembles of heterogeneous features

S. Tuarob; C. Tucker; M. Salathé; N. Ram 

Social media is emerging as a powerful source of communication, information dissemination and mining. Being colloquial and ubiquitous in nature makes it easier for users to express their opinions and preferences in a seamless, dynamic manner. Epidemic surveillance systems that utilize social media to detect the emergence of diseases have been proposed in the literature. These systems mostly employ traditional document classification techniques that represent a document with a bag of N-grams. However, such techniques are not optimal for social media where sparsity and noise are norms. The authors address the limitations posed by the traditional N-gram based methods and propose to use features that represent different semantic aspects of the data in combination with ensemble machine learning techniques to identify health-related messages in a heterogenous pool of social media data. Furthermore, the results reveal significant improvement in identifying health related social media content which can be critical in the emergence of a novel, unknown disease epidemic

Validating models for disease detection using twitter

T. Bodnar; M. Salathé 

Data mining social media has become a valuable resource for infectious disease surveillance. However, there are considerable risks associated with incorrectly predicting an epidemic. The large amount of social media data combined with the small amount of ground truth data and the general dynamics of infectious diseases present unique challenges when evaluating model performance. In this paper, we look at several methods that have been used to assess influenza prevalence using Twitter. We then validate them with tests that are designed to avoid and illustrate issues with the standard k-fold cross validation method. We also find that small modifications to the way that data are partitioned can have major effects on a model’s reported performance

The Social Maintenance of Cooperation through Hypocrisy

T. Bodnar; M. Salathé 

Cooperation is widespread in human societies, but its maintenance at the group level remains puzzling if individuals benefit from not cooperating. Explanations of the maintenance of cooperation generally assume that cooperative and non-cooperative behavior in others can be assessed and copied accurately. However, humans have a well known capacity to deceive and thus to manipulate how others assess their behavior. Here, we show that hypocrisy – claiming to be acting cooperatively while acting selfishly – can maintain social cooperation because it prevents the spread of selfish behavior. We demonstrate this effect both theoretically and experimentally. Hypocrisy allows the cooperative strategy to spread by taking credit for the success of the non-cooperative strategy

Understanding population displacements on location-based call records using road data

Z. Huang; U. Kumar; T. Bodnar; M. Salathé 

Large population displacements are usually observed after nature disasters. The best approximations to real world population movement in such a short temporal scale are the users’ movements patterns derived from the cell phone usage data. However, due to a lot of political, economic and privacy constraints, these sensitive data are not always available. On the other hand, population movements are usually observed on the underlying road network. The correlation between the cell phone users’ movement and the road network has yet been examined. The aim of this research is to compare the topological structure and the network metrics of the road network to the cell phone users’ movement network in Abidjan, Cote D’Ivoire, and to inspect the correlations of movement volume and road connectivity. A flooding scenario was assumed to inspect the responds from both the movement network and road network. Our analysis shows that the cellphone users’ movement network and the road network present significant similarities in terms of network partition. Our research also indicates that the road topology could be utilized as a proxy to approximate the population movement volume on above. We present an initial step to help the data-scarce area understand the population movement pattern from more readily-available road network data. Furthermore, our results suggest that traditional evacuation planning should consider the social perspective of population connections and periodical movements

The dynamics of health behavior sentiments on a large online social network

M. Salathé; D. Vu; S. Khandelwal; D. Hunter 

Modifiable health behaviors, a leading cause of illness and death in many countries, are often driven by individual beliefs and sentiments about health and disease. Individual behaviors affecting health outcomes are increasingly modulated by social networks, for example through the associations of like-minded individuals – homophily – or through peer influence effects. Using a statistical approach to measure the individual temporal effects of a large number of variables pertaining to social network statistics, we investigate the spread of a health sentiment towards a new vaccine on Twitter, a large online social network. We find that the effects of neighborhood size and exposure intensity are qualitatively very different depending on the type of sentiment. Generally, we find that larger numbers of opinionated neighbors inhibit the expression of sentiments. We also find that exposure to negative sentiment is contagious – by which we merely mean predictive of future negative sentiment expression – while exposure to positive sentiments is generally not. In fact, exposure to positive sentiments can even predict increased negative sentiment expression. Our results suggest that the effects of peer influence and social contagion on the dynamics of behavioral spread on social networks are strongly content-dependent

Influenza A (H7N9) and the importance of digital epidemiology

M. Salathé; C. C. Freifeld; S. R. Mekaru; A. F. Tomasulo; J. S. Brownstein 

Complex social contagion makes networks more vulnerable to disease outbreaks

E. Campbell; M. Salathé 

Social network analysis is now widely used to investigate the dynamics of infectious disease spread. Vaccination dramatically disrupts disease transmission on a contact network, and indeed, high vaccination rates can potentially halt disease transmission altogether. Here, we build on mounting evidence that health behaviors – such as vaccination, and refusal thereof – can spread across social networks through a process of complex contagion that requires social reinforcement. Using network simulations that model health behavior and infectious disease spread, we find that under otherwise identical conditions, the process by which the health behavior spreads has a very strong effect on disease outbreak dynamics. This dynamic variability results from differences in the topology within susceptible communities that arise during the health behavior spreading process, which in turn depends on the topology of the overall social network. Our findings point to the importance of health behavior spread in predicting and controlling disease outbreaks.

3D Optical Imaging of Living Cells in Microgravity : Application to Study Dynamic Changes of Cytoskeleton

Non-genetic inheritance and the patterns of antagonistic coevolution

R. Mostowy; J. Engelstädter; M. Salathé 

Antagonistic species interactions can lead to coevolutionary genotype or phenotype frequency oscillations, with important implications for ecological and evolutionary processes. However, direct empirical evidence of such oscillations is rare. The rarity of observations is generally attributed to inherent difficulties of ecological and evolutionary long-term studies, to weak or absent interaction between species, or to the absence of negative frequency-dependence.

Digital epidemiology

M. Salathé; L. Bengtsson; T. J. Bodnar; D. D. Brewer; J. S. Brownstein et al. 

Mobile, social, real-time: the ongoing revolution in the way people communicate has given rise to a new kind of epidemiology. Digital data sources, when harnessed appropriately, can provide local and timely information about disease and health dynamics in populations around the world. The rapid, unprecedented increase in the availability of relevant data from various digital sources creates considerable technical and computational challenges.

Governing the global commons with local institutions

T. Bodnar; M. Salathé 

Most problems faced by modern human society have two characteristics in common–they are tragedy-of-the-commons type of problems, and they are global problems. Tragedy-of-the-commons type of problems are those where a commonly shared resource is overexploited by free riders at the expense of everyone sharing the resource. The exploitation of global resources such as clean air and water, political stability and peace, etc. underlies many of the most pressing human problems. Punishment of free riding behavior is one of the most frequently used strategies to combat the problem, but the spatial reach of sanctioning institutions is often more limited than the spatial effects of overexploitation. Here, we analyze a general game theoretical model to assess under what circumstances sanctioning institutions with limited reach can maintain the larger commons. We find that the effect of the spatial reach has a strong effect on whether and how the commons can be maintained, and that the transitions between those outcomes are characterized by phase transitions. The latter indicates that a small change in the reach of sanctioning systems can profoundly change the way the global commons can be managed.

Measuring school contact networks using wireless sensor technology

J. Jones; M. Salathé; M. Feldman; M. Kazandjieva; J. Lee et al. 

Interaction networks shaped by social processes constitute the substrate on which various phenomena of interest to human biology occur, for example, epidemics, diffusion of health information, and the exertion of social influence related to health. Understanding the structure of network formation is thus crucial to our understanding of how relational human interactions mediate key biosocial outcomes. Research in this area has been hampered by a lack of high-quality data on the formation and structure of contact networks. Using Wireless Sensor Network (WSN) technology, measured the temporal dynamics of close-proximity interaction networks during a typical school day in a high school in the San Francisco Bay Area. Participants wore small wireless sensors which send and receive radio signals to and from other sensors nearby. This technology allowed us to collect dynamic contact network data with unparalleled precision. At a 94% coverage, we collected 762,868 CPIs at a maximal distance of 3 meters among 788 individuals. The data revealed a high density network with typical small world properties and a relatively homogenous distribution of both interaction time and interaction partners among subjects. Computer simulations of the spread of an influenza-like disease on the weighted contact graph are in good agreement with absentee data during the most recent influenza season. Analysis of targeted immunization strategies suggested that contact network data are required to design strategies that are significantly more effective than random immunization. Immunization strategies based on contact network data were most effective at high vaccination coverage

Assessing vaccination sentiments with online social media: implications for infectious disease dynamics and control

M. Salathé; S. Khandelwal 

There is great interest in the dynamics of health behaviors in social networks and how they affect collective public health outcomes, but measuring population health behaviors over time and space requires substantial resources. Here, we use publicly available data from 101,853 users of online social media collected over a time period of almost six months to measure the spatio-temporal sentiment towards a new vaccine. We validated our approach by identifying a strong correlation between sentiments expressed online and CDC-estimated vaccination rates by region. Analysis of the network of opinionated users showed that information flows more often between users who share the same sentiments – and less often between users who do not share the same sentiments – than expected by chance alone. We also found that most communities are dominated by either positive or negative sentiments towards the novel vaccine. Simulations of infectious disease transmission show that if clusters of negative vaccine sentiments lead to clusters of unprotected individuals, the likelihood of disease outbreaks is greatly increased. Online social media provide unprecedented access to data allowing for inexpensive and efficient tools to identify target areas for intervention efforts and to evaluate their effectiveness.

A high-resolution human contact network for infectious disease transmission

M. Salathé; M. Kazandjieva; J. W. Lee; P. Levis; M. W. Feldman et al. 

The most frequent infectious diseases in humans–and those with the highest potential for rapid pandemic spread–are usually transmitted via droplets during close proximity interactions (CPIs). Despite the importance of this transmission route, very little is known about the dynamic patterns of CPIs. Using wireless sensor network technology, we obtained high-resolution data of CPIs during a typical day at an American high school, permitting the reconstruction of the social network relevant for infectious disease transmission. At 94% coverage, we collected 762,868 CPIs at a maximal distance of 3 m among 788 individuals. The data revealed a high-density network with typical small-world properties and a relatively homogeneous distribution of both interaction time and interaction partners among subjects. Computer simulations of the spread of an influenza-like disease on the weighted contact graph are in good agreement with absentee data during the most recent influenza season. Analysis of targeted immunization strategies suggested that contact network data are required to design strategies that are significantly more effective than random immunization. Immunization strategies based on contact network data were most effective at high vaccination coverage.

Plague outbreaks in prairie dog populations explained by percolation thresholds of alternate host abundance

D. J. Salkeld; M. Salathé; P. Stapp; J. H. Jones 

Highly lethal pathogens (e.g., hantaviruses, hendra virus, anthrax, or plague) pose unique public-health problems, because they seem to periodically flare into outbreaks before disappearing into long quiescent phases. A key element to their possible control and eradication is being able to understand where they persist in the latent phase and how to identify the conditions that result in sporadic epidemics or epizootics. In American grasslands, plague, caused by Yersinia pestis, exemplifies this quiescent-outbreak pattern, because it sporadically erupts in epizootics that decimate prairie dog (Cynomys ludovicianus) colonies, yet the causes of outbreaks and mechanisms for interepizootic persistence of this disease are poorly understood. Using field data on prairie community ecology, flea behavior, and plague-transmission biology, we find that plague can persist in prairie-dog colonies for prolonged periods, because host movement is highly spatially constrained. The abundance of an alternate host for disease vectors, the grasshopper mouse (Onychomys leucogaster), drives plague outbreaks by increasing the connectivity of the prairie dog hosts and therefore, permitting percolation of the disease throughout the primary host population. These results offer an alternative perspective on plague’s ecology (i.e., disease transmission exacerbated by alternative hosts) and may have ramifications for plague dynamics in Asia and Africa, where a single main host has traditionally been considered to drive Yersinia ecology. Furthermore, abundance thresholds of alternate hosts may be a key phenomenon determining outbreaks of disease in many multihost-disease systems.

Experiences in measuring a human contact network for epidemiology research

M. Kazandjieva; J. Lee; M. Salathé; M. Feldman; J. Jones et al. 

This paper discusses our experience in designing and deploying a 994-node sensor network to measure the social contact network of a high school over one typical day. The system aims to capture interactions of human subjects for the study of infectious disease spread. We describe unique challenges posed by a large-scale network that is heavily affected by humans. We present techniques to address challenges such as frequent node reboots and global timestamps. The end result of the deployment is a dataset of 792 traces which can be used to calculate the school population’s contact network and the rough location where interactions occurred

On the evolution of sexual reproduction in hosts coevolving with multiple parasites

R. Mostowy; M. Salathé; R. D. Kouyos; S. Bonhoeffer 

Host-parasite coevolution has been studied extensively in the context of the evolution of sex. Although hosts typically coevolve with several parasites, most studies considered one-host/one-parasite interactions. Here, we study population-genetic models in which hosts interact with two parasites. We find that host/multiple-parasite models differ nontrivially from host/single-parasite models. Selection for sex resulting from interactions with a single parasite is often outweighed by detrimental effects due to the interaction between parasites if coinfection affects the host more severely than expected based on single infections, and/or if double infections are more common than expected based on single infections. The resulting selection against sex is caused by strong linkage-disequilibria of constant sign that arise between host loci interacting with different parasites. In contrast, if coinfection affects hosts less severely than expected and double infections are less common than expected, selection for sex due to interactions with individual parasites can now be reinforced by additional rapid linkage-disequilibrium oscillations with changing sign. Thus, our findings indicate that the presence of an additional parasite can strongly affect the evolution of sex in ways that cannot be predicted from single-parasite models, and that thus host/multiparasite models are an important extension of the Red Queen Hypothesis.

Dynamics and control of diseases in networks with community structure

M. Salathé; J. H. Jones 

The dynamics of infectious diseases spread via direct person-to-person transmission (such as influenza, smallpox, HIV/AIDS, etc.) depends on the underlying host contact network. Human contact networks exhibit strong community structure. Understanding how such community structure affects epidemics may provide insights for preventing the spread of disease between communities by changing the structure of the contact network through pharmaceutical or non-pharmaceutical interventions. We use empirical and simulated networks to investigate the spread of disease in networks with community structure. We find that community structure has a major impact on disease dynamics, and we show that in networks with strong community structure, immunization interventions targeted at individuals bridging communities are more effective than those simply targeting highly connected individuals. Because the structure of relevant contact networks is generally not known, and vaccine supply is often limited, there is great need for efficient vaccination algorithms that do not require full knowledge of the network. We developed an algorithm that acts only on locally available network information and is able to quickly identify targets for successful immunization intervention. The algorithm generally outperforms existing algorithms when vaccine supply is limited, particularly in networks with strong community structure. Understanding the spread of infectious diseases and designing optimal control strategies is a major goal of public health. Social networks show marked patterns of community structure, and our results, based on empirical and simulated data, demonstrate that community structure strongly affects disease dynamics. These results have implications for the design of control strategies.

Modelling the influence of human behaviour on the spread of infectious diseases: a review

S. Funk; M. Salathé; V. A. A. Jansen 

Human behaviour plays an important role in the spread of infectious diseases, and understanding the influence of behaviour on the spread of diseases can be key to improving control efforts. While behavioural responses to the spread of a disease have often been reported anecdotally, there has been relatively little systematic investigation into how behavioural changes can affect disease dynamics. Mathematical models for the spread of infectious diseases are an important tool for investigating and quantifying such effects, not least because the spread of a disease among humans is not amenable to direct experimental study. Here, we review recent efforts to incorporate human behaviour into disease models, and propose that such models can be broadly classified according to the type and source of information which individuals are assumed to base their behaviour on, and according to the assumed effects of such behaviour. We highlight recent advances as well as gaps in our understanding of the interplay between infectious disease dynamics and human behaviour, and suggest what kind of data taking efforts would be helpful in filling these gaps.

The role of epistasis on the evolution of recombination in host-parasite coevolution

R. D. Kouyos; M. Salathé; S. P. Otto; S. Bonhoeffer 

Antagonistic coevolution between hosts and parasites is known to affect selection on recombination in hosts. The Red Queen Hypothesis (RQH) posits that genetic shuffling is beneficial for hosts because it quickly creates resistant genotypes. Indeed, a large body of theoretical studies have shown that for many models of the genetic interaction between host and parasite, the coevolutionary dynamics of hosts and parasites generate selection for recombination or sexual reproduction. Here we investigate models in which the effect of the host on the parasite (and vice versa) depend approximately multiplicatively on the number of matched alleles. Contrary to expectation, these models generate a dynamical behavior that strongly selects against recombination/sex. We investigate this atypical behavior analytically and numerically. Specifically we show that two complementary equilibria are responsible for generating strong linkage disequilibria of opposite sign, which in turn causes strong selection against sex. The biological relevance of this finding stems from the fact that these phenomena can also be observed if hosts are attacked by two parasites that affect host fitness independently. Hence the role of the Red Queen Hypothesis in natural host parasite systems where infection by multiple parasites is the rule rather than the exception needs to be reevaluated.

Evolution of stochastic switching rates in asymmetric fitness landscapes

M. Salathé; J. Van Cleve; M. W. Feldman 

Uncertain environments pose a tremendous challenge to populations: The selective pressures imposed by the environment can change so rapidly that adaptation by mutation alone would be too slow. One solution to this problem is given by the phenomenon of stochastic phenotype switching, which causes genetically uniform populations to be phenotypically heterogenous. Stochastic phenotype switching has been observed in numerous microbial species and is generally assumed to be an adaptive bet-hedging strategy to anticipate future environmental change. We use an explicit population genetic model to investigate the evolutionary dynamics of phenotypic switching rates. We find that whether or not stochastic switching is an adaptive strategy is highly contingent upon the fitness landscape given by the changing environment. Unless selection is very strong, asymmetric fitness landscapes-where the cost of being maladapted is not identical in all environments-strongly select against stochastic switching. We further observe a threshold phenomenon that causes switching rates to be either relatively high or completely absent, but rarely intermediate. Our finding that marginal changes in selection pressures can cause fundamentally different evolutionary outcomes is important in a wide range of fields concerned with microbial bet hedging.

On the causes of selection for recombination underlying the red queen hypothesis

M. Salathé; R. D. Kouyos; S. Bonhoeffer 

The vast majority of plant and animal species reproduce sexually despite the costs associated with sexual reproduction. Genetic recombination might outweigh these costs if it helps the species escape parasite pressure by creating rare or novel genotypes, an idea known as the Red Queen hypothesis. Selection for recombination can be driven by short- and long-term effects, but the relative importance of these effects and their dependency on the parameters of an antagonistic species interaction remain unclear. We use computer simulations of a mathematical model of host-parasite coevolution to measure those effects under a wide range of parameters. We find that the real driving force underlying the Red Queen hypothesis is neither the immediate, next-generation, short-term effect nor the long-term effect but in fact a delayed short-term effect. Our results highlight the importance of differentiating clearly between immediate and delayed short-term effects when attempting to elucidate the mechanism underlying selection for recombination in the Red Queen hypothesis.

Early assessment of anxiety and behavioral response to novel swine-origin influenza A(H1N1)

J. H. Jones; M. Salathé 

Since late April, 2009, a novel influenza virus A (H1N1), generally referred to as the “swine flu,” has spread around the globe and infected hundreds of thousands of people. During the first few days after the initial outbreak in Mexico, extensive media coverage together with a high degree of uncertainty about the transmissibility and mortality rate associated with the virus caused widespread concern in the population. The spread of an infectious disease can be strongly influenced by behavioral changes (e.g., social distancing) during the early phase of an epidemic, but data on risk perception and behavioral response to a novel virus is usually collected with a substantial delay or after an epidemic has run its course.

The state of affairs in the kingdom of the Red Queen

M. Salathé; R. D. Kouyos; S. Bonhoeffer 

One of the most prominent hypotheses to explain the ubiquity of sex and recombination is based on host-parasite interactions. Under the name of the Red Queen hypothesis (RQH), it has had theoretical and empirical support since its conception, but recent theoretical work has shown that the circumstances under which the RQH works remain unclear. Here we review the current status of the theory of the RQH. We argue that recent theoretical work calls for new experimental data and an increased theoretical effort to reveal the driving force of the RQH.

The effect of opinion clustering on disease outbreaks

M. Salathé; S. Bonhoeffer 

Many high-income countries currently experience large outbreaks of vaccine-preventable diseases such as measles despite the availability of highly effective vaccines. This phenomenon lacks an explanation in countries where vaccination rates are rising on an already high level. Here, we build on the growing evidence that belief systems, rather than access to vaccines, are the primary barrier to vaccination in high-income countries, and show how a simple opinion formation process can lead to clusters of unvaccinated individuals, leading to a dramatic increase in disease outbreak probability. In particular, the effect of clustering on outbreak probabilities is strongest when the vaccination coverage is close to the level required to provide herd immunity under the assumption of random mixing. Our results based on computer simulations suggest that the current estimates of vaccination coverage necessary to avoid outbreaks of vaccine-preventable diseases might be too low.

Rapid parasite adaptation drives selection for high recombination rates

M. Salathé; R. D. Kouyos; R. R. Regoes; S. Bonhoeffer 

The Red Queen hypothesis proposes that sex is maintained through selection pressure imposed by coevolving parasites: susceptible hosts are able to escape parasite pressure by recombining their genome to create resistant offspring. However, previous theoretical studies have shown that the Red Queen typically selects against sex unless selection is strong, arguing that high rates of recombination cannot evolve when parasites are of low virulence. Here we show that under the biologically plausible assumption of a severe fitness cost for parasites that fail to infect, the Red Queen can cause selection for high recombination rates, and that the strength of virulence is largely irrelevant to the direction of selection for increased recombination rates. Strong selection on parasites and short generation times make parasites usually better adapted to their hosts than vice versa and can thus favor higher recombination rates in hosts. By demonstrating the importance of host-imposed selection on parasites, our findings resolve previously reported conflicting results.

Parasites lead to evolution of robustness against gene loss in host signaling networks

M. Salathé; O. S. Soyer 

Many biological networks can maintain their function against single gene loss. However, the evolutionary mechanisms responsible for such robustness remain unclear. Here, we demonstrate that antagonistic host-parasite interactions can act as a selective pressure driving the emergence of robustness against gene loss. Using a model of host signaling networks and simulating their coevolution with parasites that interfere with network function, we find that networks evolve both redundancy and specific architectures that allow them to maintain their response despite removal of proteins. We show that when the parasite pressure is removed, subsequent evolution can lead to loss of redundancy while architecture-based robustness is retained. Contrary to intuition, increased parasite virulence hampers evolution of robustness by limiting the generation of population level diversity in the host. However, when robustness emerges under high virulence, it tends to be stronger. These findings predict an increased presence of robustness mechanisms in biological networks operating under parasite interference. Conversely, the presence of such mechanisms could indicate current or past parasite interference.

The Red Queen and the persistence of linkage-disequilibrium oscillations in finite and infinite populations

R. D. Kouyos; M. Salathé; S. Bonhoeffer 

The Red Queen Hypothesis (RQH) suggests that the coevolutionary dynamics of host-parasite systems can generate selection for increased host recombination. Since host-parasite interactions often have a strong genetic basis, recombination between different hosts can increase the fraction of novel and potentially resistant offspring genotypes. A prerequisite for this mechanism is that host-parasite interactions generate persistent oscillations of linkage disequilibria (LD).

The evolution of complexity on the level of genes, individuals and populations

High epitope expression levels increase competition between T cells

A. Scherer; M. Salathé; S. Bonhoeffer 

Both theoretical predictions and experimental findings suggest that T cell populations can compete with each other. There is some debate on whether T cells compete for aspecific stimuli, such as access to the surface on antigen-presenting cells (APCs) or for specific stimuli, such as their cognate epitope ligand. We have developed an individual-based computer simulation model to study T cell competition. Our model shows that the expression level of foreign epitopes per APC determines whether T cell competition is mainly for specific or aspecific stimuli. Under low epitope expression, competition is mainly for the specific epitope stimuli, and, hence, different epitope-specific T cell populations coexist readily. However, if epitope expression levels are high, aspecific competition becomes more important. Such between-specificity competition can lead to competitive exclusion between different epitope-specific T cell populations. Our model allows us to delineate the circumstances that facilitate coexistence of T cells of different epitope specificity. Understanding mechanisms of T cell coexistence has important practical implications for immune therapies that require a broad immune response.

Mutation accumulation in space and the maintenance of sexual reproduction

M. Salathé; R. Salathé; P. Schmid-Hempel; S. Bonhoeffer 

The maintenance of sexual reproduction remains one of the major puzzles of evolutionary biology, since, all else being equal, an asexual mutant should have a twofold fitness advantage over the sexual wildtype. Most theories suggest that sex helps either to purge deleterious mutations, or to adapt to changing environments. Both mechanisms have their limitations if they act in isolation because they require either high genomic mutation rates or very virulent pathogens, and it is therefore often thought that they must act together to maintain sex. Typically, however, these theories have in common that they are not based on spatial processes. Here, we show that local dispersal and local competition can explain the maintenance of sexual reproduction as a means of purging deleterious mutations. Using a spatially explicit individual-based model, we find that even with reasonably low genomic mutation rates and large total population sizes, asexual clones cannot invade a sexual population. Our results demonstrate how spatial processes affect mutation accumulation such that it can fully erode the twofold benefit of asexuality faster than an asexual clone can take over a sexual population. Thus, the cost of sex is generally overestimated in models that ignore the effects of space on mutation accumulation.

The effect of multifunctionality on the rate of evolution in yeast

M. Salathé; M. Ackermann; S. Bonhoeffer 

Multifunctional genes are expected to evolve at lower rates because mutations in such genes that improve one function might often have deleterious effects on other functions. Here we tested for an association between multifunctionality and evolutionary rates in genes of Saccharomyces cerevisiae, and we find a highly significant negative correlation between the number of biological processes in which a gene is involved in and its rate of evolution. However, the magnitude of this effect is small, and the results do not support the notion that multifunctionality limits a gene’s rate of evolution.