This repository is used for coordinating all the software development activities by the ForeSITE group.
The list includes 40 tools. A complete list featuring completed and work-in-progress tools holds 55 entries. We will be updating this list as we progress.
| Type | N | |
|---|---|---|
| 3 | Epidemic Model - Scenario Modeling | 20 |
| 6 | Parameter estimation | 12 |
| 4 | Epidemic Model - Scenario Modeling & Parameter estimation | 2 |
| 1 | Decision Support tool | 1 |
| 2 | Epidemic Model - Forecasting | 1 |
| 5 | Nowcasting tool | 1 |
| 7 | Statistical Model - Forecasting | 1 |
Number of software packages by type
| Language | N |
|---|---|
| r | 20 |
| Other | 11 |
| python | 7 |
| c++ | 3 |
| anylogic | 2 |
| matlab | 2 |
| sas | 2 |
| stata | 2 |
Number of software packages by programming language.
List of existing software packages that are being used by the ForeSITE group. This list was last updated on 2024-11-26.
| N | Tool | Description |
|---|---|---|
| 1 | Airborne release of infectious pathogens simulator (R) | Estimate airborne dispersal, human exposure, and infection probabilities and timelines after a release of a quantity of infectious organisms. Scenarios(s) Modeled: Airborne release and human inhalational exposure and infection, similar the Sverdlovsk anthrax leak of 1979. (Learn more…) |
| 2 | Attention-Based Models for Snow-Water Equivalent Prediction (Python) | Transformer architectures for spatio-temporal prediction (or synthetic data generation/imputation). Scenarios(s) Modeled: Predicting the SWE value for multiple SNOTEL locations in the Western US using the Attention Models (Learn more…) GitHub:  |
| 3 | Bayesian Transmission Model (C++) | Provides estimates for critical epidemiological parameters that characterize the spread of bacterial pathogens in healthcare settings. Parameter estimated: Transmission rate (frequency-dependent or density-dependent mass action), importation probability, clearance rate (loss of colonization per colonized person per unit time), surveillance test sensitivity, surveillance test specificity, effect of covariate on transmission (multiplier in relation to overall transmission rate). (Learn more…) GitHub:  |
| 4 | Branching process outbreak simulator (R) | Quantifies risk posed by individual importers of a novel transmissible pathogen to a generic population, with intervention effects. Scenarios(s) Modeled: Novel introduction of transmissible pathogen by infected traveler, by accidentally infected laboratory worker, or similar scenario; intervention scenarios for improved detection of initial case and for delayed mitigation after ongoing outbreak is detected. (Learn more…) GitHub:  |
| 5 | Carriage duration estimation from serial testing data (R) | Estimate the duration and heterogeneity of individuals’ colonization episodes for organisms of interest. Parameter estimated: Average and distribution of clearance rate(s) across multiple candidate model forms, average (re)acquisition rate, sensitivity/specificity of testing. Estimates derived via maximum likelihood techniques. (Learn more…) |
| 6 | Competing Risks (SAS) | Estimating a competing risks mixture model, which provides both the odds ratio of one event occurring vs. another as well as a marginal estimate of the time to both respective events. Parameter estimated: Odds ratio of two different events occurring and the time to event estimate for both events. (Learn more…) |
| 7 | COVID-19 Jail Model (R) | Deterministic model of a stylized jail, with a number of potential interventions developed in collaboration with the ACLU. Scenarios(s) Modeled: Decarceration, social distancing, targeted or widespread arrest reduction, isolation. (Learn more…) GitHub:  |
| 8 | COVID-19 NCAA Football (Python) | Stochastic representation a season of on-campus football during COVID. Scenarios(s) Modeled: Estimates a football season with high or low mixing between the visiting community and the hosting college student population, and with high or low prevalence in the visitors. (Learn more…) GitHub:  |
| 9 | Covid Scenario model (R) | A flexible scenario modeling pipeline that could quickly tailor models for decision makers seeking to compare projections of epidemic trajectories and healthcare impacts from multiple intervention scenarios in different locations. Scenarios(s) Modeled: Mask, social distance, lock down, comprehensive test/isolate, fatiguing interventions (Learn more…) GitHub:  |
| 10 | CU Metapopulation (Python) | Stochastic representation of an ICU in a tertiary care academic medical center. Scenarios(s) Modeled: Staffing changes, cohorting (with some effort), decolonization, fecal transplant, antibiotic stewardship (with some modifications). Others are possible. (Learn more…) GitHub:  |
| 11 | Cumulative infection (R, SAS) | We’ve used Mortality Mapping (mMAP), an existing epidemiologic and statistical analysis approach by Lu et al. [1,2] to estimate the cumulative incidence of symptomatic COVID-19 in Veterans in the United States. mMAP is a time series deconvolution method which infers the true COVID-19 case counts from reported COVID-19 deaths. It uses symptom onset to death distribution, smoothed time series of reported COVID-19 deaths, and symptomatic case fatality rate (sCFR) to estimate the distribution of symptomatic COVID-19 cases. We focus on the adjusted version of the mMAP approach presented by Lu et al., which tries to account for the unreported COVID-19 deaths. While Lu uses excess influenza and pneumonia deaths to account for unreported COVID-19 deaths, we have extended their approach by using excess all-cause deaths. (Learn more…) |
| 12 | DEDiscover (differential evolution) (Kotlin, C++.) | Parameter estimation using differential evolution. (Learn more…) |
| 13 | Dose-response estimators for infection risk (R) | Estimate infection probability from a given dose of exposure using data sets from experimental dose-quantified exposures of humans or animals; model-based results used to perform quantitative risk assessment for potential human exposure scenarios. Parameter estimated: Per-unit-dose infection probability and associated additional parameters for more complicated dose-response models, including those that quantify dose-dependent latent/incubation period; can convert estimates to common “infectious dose” quantification metrics such as “ID-50.” Estimates derived via maximum likelihood for multiple equation-based dose-response models fit to exposure-response data. (Learn more…) |
| 14 | epiworld: Fast Agent-Based Epi Models (R, C++, Python, Webessembly) | A flexible framework for Agent-Based Models (ABM), the ‘epiworldR’ package provides methods for prototyping disease outbreaks and transmission models using a ‘C++’ backend, making it very fast. It supports multiple epidemiological models, including the Susceptible-Infected-Susceptible (SIS), Susceptible-Infected-Removed (SIR), Susceptible-Exposed-Infected-Removed (SEIR), and others, involving arbitrary mitigation policies and multiple-disease models. Users can specify infectiousness/susceptibility rates as a function of agents’ features, providing great complexity for the model dynamics. Furthermore, ‘epiworldR’ is ideal for simulation studies featuring large populations. (Learn more…) GitHub:  |
| 15 | epiworldpy: A python wrapper for epiworld (Python) | A flexible framework for Agent-Based Models (ABM), the ‘epiworldR’ package provides methods for prototyping disease outbreaks and transmission models using a ‘C++’ backend, making it very fast. It supports multiple epidemiological models, including the Susceptible-Infected-Susceptible (SIS), Susceptible-Infected-Removed (SIR), Susceptible-Exposed-Infected-Removed (SEIR), and others, involving arbitrary mitigation policies and multiple-disease models. Users can specify infectiousness/susceptibility rates as a function of agents’ features, providing great complexity for the model dynamics. Furthermore, ‘epiworldR’ is ideal for simulation studies featuring large populations. (Learn more…) GitHub:  |
| 16 | GeoCovid (R) | Shiny App for visualizing mobility data and COVID-19 cases in Buenos Aires province during March to December 2020. (Learn more…) GitHub:  |
| 17 | H1N1 Pandemic Data Estimation (TBD) | Describes an SIR model parameter estimation with an improved fit using least squares estimation over maximum likelihood estimation. Parameter estimated: Transmission rate, recovery rate, R0, number of initial infectives (Learn more…) |
| 18 | Healthcare facility network outbreak simulator (Anylogic) | Provides flexible framework for agent-based simulation of outbreaks among in-patients in a generic-but-realistic network of short/long-stay hospitals and nursing homes, as well as interventions to mitigate transmission and inter-facility spread. Scenarios(s) Modeled: Initial invasion of new pathogen / strain and/or endemic pathogens. Intervention scenarios for patient-targeted efforts including surveillance and isolation (contact precautions), vaccination, and anti-microbial drug treatment. (Learn more…) |
| 19 | Healthcare transmission epidemiology estimation at equilibrium (R) | Estimate the extent of patient-to-patient transmission and other quantities using admission prevalence and cross-sectional point-prevalence data for carriage of organisms among healthcare facility in-patients. Parameter estimated: Patient-to-patient transmission rate, facility reproduction number, rate of progression to infection, intervention effects if applicable. Estimates derived via solving differential equations at equilibrium and facility reproduction number equations. (Learn more…) |
| 20 | Herpes Vaccine Model (Matlab) | An ODE-based 6-compartment SI model with asymptomatic carriers and vaccinated classes to forecast the number of infections prevented with imperfect vaccination and infections prevented per vaccine administered. Parameter estimated: Highly correlated parameters determined using LHS PRCC. Scenarios(s) Modeled: Prophylactic (pre-exposure) vaccines, Therapeutic (post-exposure) vaccines, Parameter(s) Estimated: Highly correlated parameters determined using LHS PRCC (Learn more…) |
| 21 | HIV End Stage Kidney Disease Epidemic Model (Coded in Matlab, Berkeley Madonna, or Excel.) | An ODE-based model to predict future disease prevalence given the impact of drug therapy levels and its mechanism of action (i.e., stopping progression to disease or reducing deaths). Parameter estimated: Growth rate, death rate, drug efficacy to block disease progression. Scenarios(s) Modeled: Rebound in prevalence due to insufficient therapy (e.g., like in Paxlovid rebound). (Learn more…) |
| 22 | Household epidemiology estimation from antibody data (R) | Simultaneously estimate household importation risk, household secondary attack rate, and associated variability / heterogeneities using point-prevalence data for antibody positivity among household members. Parameter estimated: Household importation probability, household secondary attack rate, sensitivity / specificity of data for indicating prior infection, variability (dispersion) in transmissibility, and heterogeneities in those quantities over age and/or other identifiable characteristics of individuals and households. Estimates derived via maximum likelihood for final household outbreak size equations. (Learn more…) |
| 23 | Household transmission simulator (R) | Simulate household importation and within-household transmission across a population of households, and intervention effects, e.g. vaccination. Scenarios(s) Modeled: Household importation and transmission with variability in transmissibility and heterogeneity across age groups and vaccination status. (Learn more…) |
| 24 | Impact of school opening model on SARS-CoV-2 community incidence and mortality (Stata) | The goal of this statistical analysis was to compare SARS-CoV-2 community incidence and mortality rates across schools that opened in in-person, remote, or hybrid mode. Measure(s) of Economic Impact: Infections and mortality (Learn more…) |
| 25 | Long-term acute care hospital (LTACH) intervention model (R) | Provides analytic framework for investigating patient-focused interventions to prevent transmission in high-risk healthcare facilities and their direct and population-level health impacts and threshold effects. Scenarios(s) Modeled: Intervention scenarios for surveillance and isolation (contact precautions), decolonization drugs, infection treatment, horizontal transmission reduction, and length of stay reduction. (Learn more…) |
| 26 | Monolix (Monolix can be used either via a graphical user interface (GUI) or a command-line interface (CLI) for powerful scripting. R package available) | Parameter estimation in non-linear mixed effect models. Parameter estimated: Any. (Learn more…) |
| 27 | Multitask Model to Forecast Patient’s Next Unit and Remaining Length of Stay (Python) | Length of Stay prediction and unit transfer prediction tool. Scenarios(s) Modeled: In this multi-task learning model, we train the model to predict:Task-1: patient’s unit label in next time step. Task-2: patient’s remaining length of stay.The model inputs get updated on a daily basis and generates prediction accordingly. (Learn more…) GitHub:  |
| 28 | Net financial impact of halting elective admissions (Stata) | The goal of this model was to estimate the net financial impact to the hospital of halting elective admissions due to a COVID-19 surge. Measure(s) of Economic Impact: Hospital profit (Learn more…) |
| 29 | Nowcasting (TBD) | Given the time lag between laboratory testing for disease and eventual reporting, we’ve developed a model which consumes data from both community and VA to predict “now” how many positive tests for a disease (COVID-19 in this context) will eventually be reported as positive. A VA HSR&D abstract is available. (Learn more…) |
| 30 | PhenoMapper: an interactive toolbox for the visual exploration of phenomics data (PhenoMapper extension) | TDA tool that can possibly help us to view some trajectory data. Scenarios(s) Modeled: We demonstrate the utility of this new tool on real-world plant (e.g., maize) phenomics datasets. (Learn more…) GitHub:  |
| 31 | POMP (Aaron King’s tool, U. of Michigan) (R) | Analysis of time series data based on stochastic dynamical systems models (Learn more…) GitHub:  |
| 32 | pomp R code for the estimation of unknown parameters in a two-population SEIR model from data collected in Whitman County, Washington (R) | Build a two-population SEIR model using the pomp coding framework in R and estimate the unknown parameters in this model using the PMCMC algorithm. Parameter estimated: Transmission parameters (transmission within each population and transmission across the two-populations), testing probability, and overdispersion parameter. (Learn more…) |
| 33 | Post-acute care model (R) | A basic model to plan for the post-acute care needs following a hospital surge (Learn more…) GitHub:  |
| 34 | quadkeyr (Python) | An R package designed to create raster images from QuadKey-identified data (Microsoft Bing Maps Tile System), aiming to simplify the analysis of Facebook mobility data. (Learn more…) GitHub:  |
| 35 | School outbreak simulator (Anylogic) | Agent-based simulation of person-to-person transmission at schools over data-based dynamic proximity contact networks. Scenarios(s) Modeled: Transmission modeled during school hours; can model interventions affecting the contact network and school attendance behavior. (Learn more…) |
| 36 | Serfling (JMP) | Classic Serfling-type regression model for seasonal diseases. Can be used to estimate both excess risk (based on breaching a threshold), or estimate the peak week and other parameters of a yearly epidemic curve. (Learn more…) |
| 37 | Statistical Alerts (time series) (R) | We have supported an Operations system for processing time series for “aberrations” (aka, “alerts”, “potential outbreaks”) for 8+ years in VA BASIC. During this time, we’ve integrated several among the suite of algorithms available in the R package titled “surveillance”. This includes several classes of numerical methods, of which some are window-based, control chart-based or trained linear models. A brief presentation on this is attached here. (Learn more…) |
| 38 | Transmission estimation for outbreak clusters (R) | Provides estimates for individual-level transmission distribution (mean and variance) based on data from prior isolated outbreak clusters. Useful for quantifying the risk of superspreading events and the risk that future outbreaks will exceed a given threshold.Parameter estimated: Basic reproduction number; dispersion parameter quantifying variance in transmission; extensions estimate changes in reproduction number by generation, e.g. post-control reproduction number. Estimated via maximum likelihood using final size distribution equations for branching process models. (Learn more…) |
| 39 | Vaccine Equity (R) | A tool for exploring the effects of varying rates of vaccination among disparate socio-economic groups (Learn more…) GitHub:  |
| 40 | Vaccine Hesitancy Model (Matlab, R, and Mathematica.) | An ODE (ordinary differential equation)-based SVEIR model with 3 compartments for symptoms (mild, moderate, severe) that can quantify the tradeoffs between vaccine efficacy and vaccine hesitancy on reducing the disease burden. Scenarios(s) Modeled: R0, cumulative infections, and deaths due to varying values of transmission rate, vaccine coverage rate, vaccine efficacy, and vaccine hesitancy. The vaccine efficacy required to compensate for different levels of vaccine hesitancy. (Learn more…) |
