The Biocomplexity Institute has been supporting the U.S. federal government through major epidemics since 2005. Our research helped in planning for major outbreaks of H5N1, H1N1, MERS, Ebola, and Zika. Here is a timeline of our work.
Since early January 2020, Biocomplexity Institute researchers have been working with federal authorities to address questions like the risk of importation of COVID-19 to the United States and other countries; location of likely clusters within the United States; and estimation of various disease progression parameters.
Our team also analyzed a variety of counterfactual and what-if scenarios and considered the impact of social interventions. We conduct real-time simulations to forecast disease spread to ascertain the specifics: when, where, and the intensity of the outbreak. Based on these results, we identify the most efficient ways to estimate and allocate scarce medical resources. To this end, we have developed support tools and dashboards to help government officials, healthcare providers, and the general public to better evaluate the impact of the pandemic.
COVID-19 and the Research Community
During the Ebola crisis in West Africa in 2014, simulations and modeling of disease outbreaks were found to be an integral tool in fighting the epidemic. Mathematical and computational models allow researchers to explore the scope of an emerging outbreak by forecasting how severe the impact could be if the disease goes unchecked. Models can be designed to overcome limitations caused by sparse information or missing data, and can be adapted as more information is known. A range of epidemic forecasts, including best- and worst-case scenarios, can be especially useful to policymakers so they can better gauge how to budget and distribute limited resources. Additionally, analyzing results of multiple models, and assessing the differences between them, can lead to a better understanding of the disease parameters and guide revisions that lead to better forecasts. Modeling can also help epidemiologists study the effectiveness of different interventions, quickly and cheaply evaluating which interventions (and combinations of interventions) are likely to be most effective in resolving the outbreak and reducing disease spread.
Although modeling can be useful for predicting how severe an epidemic may be and the strategies to combat it, simulations can also be useful for understanding why a disease has emerged where it has, and why at this time. Model-based risk assessment of emerging diseases can help determine which outbreaks are likely to become epidemics, even in the absence of prior related outbreaks. This deductive reasoning approach can be used in coordination with inductive reasoning often employed by epidemiologists; as the disease progresses, forecasted data can be compared to the observed data, and the models can be revised accordingly; this kind of coordinated cross-analysis can help resolve uncertainties in the disease progression.
Models can also capitalize on data from diverse sources, such as cultural mores, social networks, transportation, weather, and the interface between the human and animal worlds. Taken individually, these data elements may not appear to have a strong correlation with the disease outbreak, but pulled together, these various data elements may draw a picture that provides a better understanding of why here and why now. Understanding how humans connect and interact with the world around them can help overcome deficiencies in information known about the disease itself.
In the case of COVID-19, there are certain features of coronaviruses in general that can be considered when forecasting future spread. However, there was some delay between the emergence of the disease and the announcement to the epidemiological community. This has been complicated by language barriers as well as political obstacles. The available data is simultaneously sparse but vast; although we didn’t have detailed information at first, now we have much more information from China and the other affected countries, as well as data from cruise ships, quarantine areas, and more. Limitations in screening techniques have caused people to be misdiagnosed. Also, fears of being quarantined have led some people to take steps to hide their illness from screeners, so prevalence is likely to be much higher than reported. All of these factors make it hard to assess the epidemic by traditional means, but modeling and simulation can help overcome these limitations and better inform policymakers and health workers who have to address this crisis.
For more on why modeling is a useful tool for epidemic research, see this article.
The most important strategy you can employ to protect yourself and your loved ones from COVID-19 is to prepare and not panic. While staying up to date on global and national cases is useful, knowing how your local community and hospitals are planning for travel-related cases and sustained transmissions will be more relevant to your immediate situation. You should also listen to updates and advice from local health authorities, government officials, and federal agencies, such as the Centers for Disease Control and Prevention (CDC).
Additional guidelines can be found on the CDC's Website.
As we continue to support the planning and response efforts during the COVID-19 pandemic our research team has prepared a visualization tool that provides a unique way of examining data curated by different data sources.
The United States
This video shows how the COVID-19 confirmed number changes for states in the U.S. from January 22, 2020 to early October. On the right, each state is color-coded by its confirmed number; data for U.S. and the world, i.e., all regions, along with a chart on the left for the world's daily count.
This video shows how the COVID-19 confirmed number changes for the entire world from January 22, 2020 to early October. On the right, each country (state/province in U.S./China) is color-coded by its confirmed number; data for U.S. and the world, i.e., all regions, along with a chart on the left for the world's daily count.
This video shows how the COVID-19 confirmed number changes for counties in Virginia from March 5, 2020 to early October. On the right, each county is color-coded by its confirmed number; data for the world, i.e., all regions, Virginia and top 10 counties in Virginia are shown on the left.
Our research team works hard to curate data related to all aspects of COVID-19. We would like to share this data with the scientific community, and will update content with links to those data sets throughout our response support efforts.
Our dashboard data is currently available in CSV format by choosing the blue "Download All" button on the upper left side of the dashboard. Click the "FAQ" button to learn more about our methodologies, citation requests, and disclaimers.
The contacts analyzed here are derived from a synthetic population of the United States in which each person is assigned an activity template based on American Community Survey data and locations for each activity are assigned based on location types, sizes, and distance from residence or catchment region. Contacts are inferred between people present at a location simultaneously. For locations with many occupants, a subgraph is constructed that interpolates between a minimum and maximum number of contacts, depending on the total number of occupants. Activities are assigned differently for weekdays and weekends. Hence the contact networks also differ. In addition, we analyzed an activity-independent network the includes only household contacts.
This archive contains mobility data made public by Baidu and scraped from their qiangxi.baidu.com web site in February and March of 2020. We have reformatted the data into a more easily computable form, comma-separated value (CSV) files providing the full origin-destination matrix for each day. We are publishing this reformatted version for research purposes under Article 22 of the Copyright Law of the People's Republic of China. We make no representation as to the suitability of the data for any purpose, but nevertheless hope that it may be useful for researchers trying to calibrate models of COVID-19. We wish to thank Baidu especially for making these valuable data available and encourage them to continue to do so.
Preprints and Perspective Articles
Peddireddy A, Xie D, Patil P, Wilson ML, Machi D, Venkatramanan S, Klahn B, Porebski P, Bhattacharya P, Dumbre S, Raymond E, Marathe M. From 5Vs to 6Cs: Operationalizing Epidemic Data Management with COVID-19 Surveillance. medRxiv. 2020 Oct 31.
Wang L, Adiga A, Venkatramanan S, Chen J, Lewis B, Marathe M. Examining Deep Learning Models with Multiple Data Sources for COVID-19 Forecasting. arXiv:2010.14491. 2020 Oct 27.
Adiga A, Dubhashi D, Lewis B, Marathe M, Venkatramanan S, Vullikanti A. Models for COVID-19 Pandemic: A Comparative Analysis. arXiv:2009.10014. 2020 Sep 21.
Eubank S, Eckstrand I, Lewis B, Venkatramanan S, Marathe M, Barrett CL. Commentary on Ferguson, et al., “Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand”. Bulletin of Mathematical Biology. 2020 Apr;82:1-7.
Wu N, Ben X, Green B, Rough K, Venkatramanan S, Marathe M, Eastham P, Sadilek A, O'Banion S. Predicting Onset of COVID-19 with Mobility-Augmented SEIR Model. medRxiv. 2020 Oct 5.
- Marathe M (2020) High Performance Simulations to Support Real-Time COVID-19 Response - ACM SIGSIM PADS, 16 June 20
- Venkatramanan S (2020) Rural Populations and Infectious Disease Transmission: Implications for COVID-19 - COVID-19 CHHS Webinar Series, GMU 2 June 20
- Venkatramanan S (2020) Computational modeling and data strategies for predicting and responding to pandemics - Intelligent Health Inspired Online Summit, 27 May 20
- Marathe M (2020) Computational and Statistical Tools to Control a Pandemic - Theoretically Speaking Series, Simons Institute 11 May 20
- Marathe M (2020) Tackling the COVID-19 crisis at the University of Virginia - 11 May 20
- Adiga A, Venkatramanan S, Schlitt J, Peddireddy A, Dickerman A, Bura A, Warren A, Klahn B, Mao C, Xie D, Machi D, Raymond E, Meng F, Barrow G, Mortveit H, Chen J, Walke J, Goldstein J, Wilson M, Orr M, Porebski P, Telionis P, Beckman R, Hoops S, Eubank S, Baek Y, Lewis B, Marathe M, Barrett C (2020) Estimating Global direct importation risk for COVID-19.
2014 – 2017 Presentations
- Mathematical and Computational Foundations of Infectious Disease Epidemiology - A 4-hour tutorial providing an overview of the state-of-the-art in mathematical and computational epidemiology, which have typically not been studied from a multi-disciplinary perspective. International Conference on Systems Biology (ICSB). Blacksburg, VA, August 2017. Presented jointly by M Marathe, A Vullikanti, and B Lewis.
- Computational Epidemiology and Public Health Policy Planning - A 4-hour tutorial that provides an overview of the state-of-the-art in computational epidemiology from a multi-disciplinary perspective. 30th Annual Conference on Artificial Intelligence (AAAI). Phoenix, AZ, 13 Feb 2016. Presented jointly by M Marathe, A Vullikanti, and N Ramakrishnan.
- Computational Epidemiology - A 3-hour course on this state-of-the-art multi-disciplinary research area, including data mining, machine learning, high-performance computing, and theoretical computer science, as well as math, economics, and statistics. Knowledge Discovery and Data Mining Conference (ACM KDD) 2014. New York, NY, August 2014. Presented jointly by M Marathe, N Ramakrishnan, and A Vullikanti.
UVA and COVID-19 In The News
Prince William loses 3 more residents to COVID-19, adds 87 new cases - Prince William Times 10 Nov 20
Weekly COVID-19 cases in Virginia could double by December, new model shows - INSIDENOVA 24 Oct 20
Could Virginia be heading for a new COVID-19 case peak? - Virginia Mercury 21 Oct 20
UVA COVID-19 model predicts coronavirus cases in Virginia could peak by Thanksgiving - ABC 8 News 21 Oct 20
Hospital beds at UVA stable with students returning to Grounds - CBS19 6 Sept 20
Model Predicts 187K Total Coronavirus Cases In Virginia By Thanksgiving - Patch.com 31 Aug 20
US Campuses try chiding and suspending students to stay open - Times Higher Education 25 Aug 20
Coronavirus in DC, Maryland, Virginia: What to Know on Aug. 18 - NBC Washington 18 Aug 20
Learning in the Midst of the COVID-19 Pandemic: Researchers, Clinicians Weigh In - UVA Today 14 Aug 20
UVA modelers forecast COVID-19 cases to peak end of September at more than double the current weekly rate - The Roanoke Times 14 Aug 20
UVA Biocomplexity Institute tries to solve puzzle of projecting global pandemic - S&P Global Market Intelligence 4 Aug 20
Some economic activity in Virginia nearly back to prepandemic levels, according to cellphone data study - Washington Business Journal 3 Aug 20
UVA researchers say state has avoided more than 800,000 additional COVID-19 cases but majority 'still vulnerable to infection' - Richmond Times-Dispatch 31 July 20
UVA COVID-19 modeling predicts steep spike in virus cases in September - NBC29 28 July 20
Northam warns of restrictions as COVID-19 cases near previous peak - The Roanoke Times 27 July 20
Some Virginia's health districts are seeing a surge in COVID-19, sounding the alarm for public health experts - Virginia Mercury 27 July 20
Researchers Present Algorithm that may Predict the Covid-19 Pandemic's Immediate Future - Yahoo! News 7 July 20
COVID-19 Drives the Need for SuperComputing Resources in Research - HPC Wire 6 July 20
Can an Algorithm Predict the Pandemic's Next Moves? - The New York Times 2 July 20
COVID-19 Cases in Virginia Remain Stable - Alexandria Living Magazine 1 July 20
COVID-19 CHHS Webinar Series: Rural Populations and Infectious Disease Transmission: Implications for COVID-19 - College of Health and Human Services, GMU 24 June 20
Virginia's COVID metrics are trending down. Here's why the state isn't reopening faster - Virginia Mercury 22 June 20
COVID-19 cases could rise as Virginians move toward Phase 3 reopening - The Roanoke Times 19 June 20
As temps soar this summer, here are some useful tips for wearing a mask and trying to stay cool - AccuWeather 18 June 20
Flattening the curve with Dr. Bryan Lewis and Dr. Srinivasan Venkatramanan - Mini Med Podcast 17 June 20
Coronavirus crisis is now 2nd-worst pandemic in US history - AccuWeather 17 June 20
UPDATE: Fauquier County records five new cases Wednesday morning, for total of 398 - Fauquier Times 17 June 20
UVA doctors and students launch 12-part COVID-19 podcast series - News-medical.net 17 June 20
UVA medical students, physicians, develop COVID-19 podcast - CBS19 News 17 June 20
UVA to Lead Study of How Big Data Can Battle Future Pandemics - wvtf.org 1 June 20
Virginia could peak at more than 5,000 new COVID-19 cases a day this summer, UVA model projects - Richmond-Times Dispatch 28 May 20
Indian IT helps global healthcare firms fight Covid-19 - Economic Times 27 May 20
Low- and middle-income countries face up to COVID-19 - Nature Medicine 21 May 20
Models Predict Uptick in Cases as Virginia Reopens - VPM.org 21 May 20
Infectious disease models aren't crystal balls but are useful tools in Florida's fight against COVID-19, modelers say - Naples Daily News 21 May 20
The contact tracing debate: Do we need to sacrifice personal data privacy to get back to business? - Silicon Angle 19 May 20
Towards Pandemic-proof Cities - Direct Industry E-magazine 19 May 20
Geographic information is crucial to modern healthcare - Geospatial World 18 May 20
U.S. researchers map COVID-19 cases to predict future needs - US Embassy in Georgia 15 May 20
The Software That's Powering All the Coronavirus Dashboards - Bloomberg Businessweek 12 May 20
Outlook into the Worldwide Digital Twin Industry to 2035 - Business Wire 8 May 20
Stanford crowdsourcing collects county-level social-distancing data - The Perry News 6 May 20
The Tricky Math Behind Coronavirus Death Predictions - The Wall Street Journal 5 May 20
Inside the Numbers: COVID-19 hitting Virginia seniors hardest - Augusta Free Press 4 May 20
Virginia holding drive-through testing in Giles County - Bluefield Daily Telegraph 3 May 20
How COVID-19's US fatalities compare to America's worst flu seasons - AccuWeather 1 May 20
How prepared are we for the next pandemic? - Healthworld.com 30 Apr 20
Lifeline: Nurses, doctors and staff at local hospitals prepare for COVID-19 long haul - Charlottesville Tomorrow 30 Apr 20
'We're saving lives': Science says social distancing is working against COVID-19 - Delmarva Now 29 Apr 20
A record-setting flu season's end doesn't ease COVID-19 worries - AccuWeather 28 Apr 20
Mapping the curve: How GIS is helping the Covid-19 response - Geographical 27 Apr 20
Will COVID-19 Intensify the Fault Lines of India's Already Unequal Society? - The Wire 27 Apr 20
Coronavirus cases in Virginia rise to 12,970 on Sunday - WHSV 26 Apr 20
COVID-19, by the numbers - The Roanoke Times 25 Apr 20
University creates new committee tasked with providing recommendations on the fall 2020 semester - The Cavalier Daily 24 Apr 20
Crowdsourcing site seeks to predict efficacy of social distancing - The Stanford Daily 23 Apr 20
UVA and RAND Corporation unveil impact modeling of COVID-19 data - Henry County Enterprise 23 Apr 20
Everything You Need to Know About Your Smartphone's New COVID-19 Tracker - UVA Today 22 Apr 20
COVID-19 testing, the key to lifting public restrictions, stalls in Virginia - The Daily Progress 18 Apr 20
Editorial: Are we there yet? No. - The Roanoke Times 16 Apr 20
5 burning questions about tech efforts to track Covid-19 cases - STAT 15 Apr 20
Restrictions have helped Virginia, but COVID-19 rates will spike if they're relaxed too soon, researchers say - The Roanoke Times 14 Apr 20
Fresh Markets Tells Shoppers to Wear Masks - Alexandria Living 14 Apr 20
Coronavirus Modeling Provides Mixed Bag of Data for Virginia - US News & World Report 13 Apr 20
Bicomplexity Institute Researchers Develop COVID-19 Projection Model for Virginia - UVA Today 13 Apr 20
UVA researchers project mid-August peak for new COVID-19 cases in Virginia - Richmond Times-Dispatch 13 Apr 20
Northam introduces new Virginia-specific COVID-19 model from UVA - Richmond Times-Dispatch 13 Apr 20
New Model Shows Virginia COVID-19 Surge in Summer if Restrictions Lifted Too Soon - NBC Washington 13 Apr 20
Easing stay-at-home order could result in mid-summer COVID-19 peak: UVA projection - Inside NoVA 13 Apr 20
UVA's coronavirus model projects Virginia will have enough hospital capacity in next few months - The Virginian Pilot 13 Apr 20
New models to show when COVID-19 might peak in Virginia - WTVR 13 Apr 20
Crowdsourcing site collects county-level policy data to inform decisions about easing social-distancing - Medical Xpress 13 Apr 20
Researchers Simulate COVID-19 Interventions on PSC's Bridges Supercomputing Platform - HPC Wire 13 Apr 20
Tata Consultancy Services, Cognizant Technology Solutions, Persistent Systems bring tech tools to Covid-19 battle - The Economic Times 9 Apr 20
Can Urban and Rural Americans Get on the Same Page About COVID-19? - Greater Good 8 Apr 20
Social distancing plays a role in flu-like illness activity in the US - AccuWeather 3 Apr 20
Can a smartphone app help us turn the coronavirus corner and go outside again? - Bulletin of the Atomic Scientists 2 Apr 20
Is the US ready for how the coronavirus may impact the next flu season? - AccuWeather 1 Apr 20
Digital coronavirus maps helping public understand the pandemic better - TechRepublic 31 Mar 20
5 ways maps can help communities respond to COVID-19 - StateScoop.com 31 Mar 20
Virginia Coronavirus Stats Only Capture 'Tip of the Iceberg' - VPM NPR News 31 Mar 20
ICMR Study Suggests Its Testing Strategy Was Flawed, Airport Screening a Miss - The Wire: Science 28 Mar 20
Influenza-like activity soars again during historic flu season amid escalating pandemic - AccuWeather 27 Mar 20
Biocomplexity Institute wins $10M Grant to Thwart Future Pandemics - UVA Today 26 Mar 20
UVA Biocomplexity Institute receives big grant - CBS19 26 Mar 20
The new coronavirus has caused unprecedented flu-like activity in the US - AccuWeather 25 Mar 20
Modeling the Spread of Epidemics - The Pragati Podcast 25 Mar 20
Digital Twins of Cities Simulate Coronavirus Public Policy Impacts - Enterprise AI 25 Mar 20
AI Platform Aims to Help Policy Makers Calibrate Virus Response - Wall Street Journal 24 Mar 20
Why some say they're comfortable exercising outdoors during the COVID-19 pandemic - AccuWeather 24 Mar 20
Virginia health officials say they can test 1,000 people for COVID-19. Experts say that's nowhere near enough. - Prince William Times 21 Mar 20
COVID-19 disruptions could last for months, researcher tells AccuWeather - AccuWeather 21 Mar 20
The impact of the new coronavirus could lead to the longest flu season in decades - AccuWeather 20 Mar 20
Brazil's densely packed favelas brace for coronovirus: 'It will kill a lot of people.' - Washington Post 20 Mar 20
Daily Coronavirus Briefing: Global death toll surges past 10,000 after most devastating day yet - Texas Border Business 20 Mar 20
Disease Experts Weigh In On Crowded Partying at Florida Beaches - Healthy Holistic Living 19 Mar 20
Clearwater draws line in sand: No admittance - Tampa Bay Times 19 Mar 20
Pinellas beaches called coronavirus risk; sheriff demands social distancing - Tampa Bay Times 18 Mar 20
Clearwater Beach will close for 2 weeks starting Monday, breaking ranks with Pinellas - Tampa Bay Times 18 Mar 20
As Long as You're Social Distancing, You Might as Well Do It Right-Here's How - Real Simple 17 Mar 20
Coronavirus Florida: Officials across Pinellas County agree to keep beaches open - Tampa Bay Times 17 Mar 20
Clearwater closes pier amid coronavirus concerns - Tampa Bay Times 17 Mar 20
The new coronavirus makes an extremely active flu season worse - Yahoo! Finance 16 Mar 20
ESRI sets up COVID-19 GIS Hub - Planetizen 13 Mar 20
Could the new coronavirus lead to a prolonged flu season? - AccuWeather 11 Mar 20
Experts warn the coronavirus could be the 'Hurricane Sandy of epidemics' - AccuWeather 4 Mar 20
World Health Organization Reports Global Mortality Rate for COVID-19 is 3.4% - The Chattanoogan.com 4 Mar 20
We Are Ignoring One Obvious Way to Fight the Coronavirus - The New York Times Editorial Board 3 Mar 20
UVA researchers provide data tracking the coronavirus - CBS19 27 Feb 20
UVA hosts talk on coronavirus’ potential impacts - CBS19 24 Feb 20