• Software

    The Genome-based Incidence Estimation Pipeline GInPipe is a fully automated pipeline that infers the incidence trajectory of a respiratory viral pandemic, like SARS-CoV-2, from a collection of time-stamped viral sequences, which are typically available from public databases such as GISAID.

    In brief: Viral sequence data is placed into redundant temporal bins. For each bin, a parameter is inferred that correlates with the incidence of infections. GInPipe then smoothes over all derived parameters and reconstructs continuous trajectory of the infection incidence. Using additional data on reported cases, this allows, among other things, to assess the level of case underreporting and to assess how underreporting is affected by policy- or behavioural changes.

    Pipeline, source code and documentation: GInPipe


    • Rapid incidence estimation from SARS-CoV-2 genomes reveals decreased case detection in Europe during summer 2020. MR Smith, M Trofimova, A Weber, Y Duport, D Kühnert and M von Kleist
      Nature Communications , 12, 6009, 2021

    The CovidStrategyCalculator is a web-based graphical interface that allows to assess arbitrary non-pharmaceutical intervention (NPI) strategies to reduce the spread of SARS-CoV-2. Those NPIs include testing regimen and quarantine/isolation periods depending on the time and occasion of SARS-CoV-2 exposure. Moreover, the tool allows to estimate the current SARS-CoV-2 prevalence based on an incidence history and a proportion of undetected cases.

    CovidStrategyCalculator essentially implements a stochastic within-host model of SARS-CoV-2 dynamics that has been fitted to clinical data and allows to estimate the probabilities of being infectious, symptomatic and  that the virus to be detectable by polymerase chain reaction (PCR) or (antigen-based) rapid diagnostic testing (RDT) at every time instance post infection. Running CovidStrategyCalculator allows to quantify the risk reduction (risk of transmission relative to an unmitigated scenario) for an user-defined NPI. The internal model is solved algebraically, so that the tool can be run within a modern browser instantly, executing in milliseconds. CovidStrategyCalculator is implemented in C++ and compiled with Qt for use within any web-browser.

    Tool (web version): https://covidstrategycalculator.github.io/

    Source code, documentation and offline software: https://github.com/CovidStrategyCalculator/COVIDStrategyCalculator


    • COVIDStrategyCalculator: A software to assess testing- and quarantine strategies for incoming travelers, contact person management and de-isolation W. van der Toorn, DY Oh, M. von Kleist, Patterns (Cell press) , 2, 100264, 2021 (online)
    • An intra-host SARS-CoV-2 dynamics model to assess testing- and quarantine strategies for incoming travelers, contact person management and de-isolation W. van der Toorn, DY Oh, D Bourquain, J Michel, E Krause, A Nitsche, M. von Kleist, Patterns (Cell press) , 2, 100262, 2021 (online)


    Mutational Interference Mapping Experiment Analysis Tool (MIMEAnTo) is a user-friendly cross-platform software to analyze data generated by Mutational Interference Mapping Experiments (MIME).

    RNA has long been believed to mainly serve as a blueprint for proteins (in the form of mRNA). However, a large diversity of so called non-coding RNAs have been discovered to regulate virtually all cellular processes. We have recently developed MIME to characterize functional domains and structure-function-relations in RNA. In MIME, target RNAs are randomly mutated, selected by function, physically separated and sequenced using next-generation sequencing (NGS). The software MIMEAnTo then allows the identification of domains and structures in non-coding RNA which are important for its function.

    Availability: https://github.com/maureensmith

    Manual (Version 1.1), Technical & mathematical details

    Platforms: Linux, Windows, OsX


    • Mutational interference mapping experiment (MIME) for studying RNA structure and function. Redmond P Smyth*, Laurence Despons, Gong Huili, Serena Bernacchi, Marcel Hijnen, Johnson Mak, Fabrice Jossinet, Li Weixi, Jean-Christophe Paillart, Max von Kleist* & Roland Marquet*, Nature Methods 12, 866–872 (2015)
    • MIMEAnTo -Profiling functional RNA in Mutational Interference Mapping Experiments. Maureen Smith, Redmond P Smyth, Roland Marquet and Max von Kleist, Bioinformatics, 32, 3369-3370, (2016) link
    • In cell mutational interference mapping experiment (in cell MIME) identifies the 5′ polyadenylation signal as a dual regulator of HIV-1 genomic RNA production and packaging. Smyth RP, Smith MR, Jousset AC, Despons L, Laumond G, Decoville T, Cattenoz P, Moog C, Jossinet F, Mougel M, Paillart JC, von Kleist M, Marquet R. Nucleic Acids Res. 46(9):e57. (2018) link

    PrEP predictor is a easy-to-use graphical user interface which predicts the utility of an antiviral for HIV prophylaxis. The software can be used to screen antivirals based on their prophylactic efficacy.

    Availability :

    Installer for Windows

    Executable for Linux

    Matlab Codes for GUI

    A brief manual

    Requirement : During the installation of PrEP predictor the installer will prompt the user to download the Matlab runtime environment. The user need to permit the download.


    • Mechanistic framework predicts drug-class specific utility of antiretrovirals for HIV prophylaxis.
      Duwal S, Dickinson L, Khoo S, von Kleist M. PLoS Comput Biol. 15(1):e1006740 (2019). link

    TOP (Treatment Optimiser) is a Matlab toolbox that implements algorithms to solve optimal control problems in the context of treatment optimization.

    Finding optimal treatment strategies is a very important and non-trivial problem. A policy maker has to take into account a number of factors such as  the health status of the patient, resource constraints (e.g. monetary) and constraints on the design of a treatment strategy among others. In our work (Duwal et al. 2015), we presented and compared two treatment paradigms: diagnostic-guided and a pro-active treatment strategies exemplified for controlling HIV-1 replication in the light of resource constraints and drug resistance evolution. A diagnostic-guided strategy tailors treatment decisions on an individual basis guided by infrequent and possibly costly diagnostics. On the other hand, a pro-active strategy suggests treatment decisions based on experience and projected outcomes. The latter allows switching treatments before drug resistance is detectable, in contrast to a diagnostic-guided strategy. However, pro-actively switching treatments may also lead to unnecessary treatment changes in some patients where drug resistance may not have developed.

    Mathematically,  a diagnostic-guided strategy can be formulated as a closed-loop optimal control problem and the optimal solution can be efficiently found using dynamical programming, e.g. as in the policy iteration algorithm (Winkelmann et al. 2014). A pro-active strategy can be described as an open-loop optimal control problem. We developed an efficient dynamic programming algorithm based on a branch-and-bound technique (Duwal et al. 2015) allowing to solve this optimization problem efficiently (and exact).

    Availability :

    Github: https://github.com/SulavDuwal/OptimalTreatmentStrategies

    Matlab file exchange: http://de.mathworks.com/matlabcentral/fileexchange/59202-sulavduwal-optimaltreatmentstrategies

    Requirement: Matlab

    Recommendation for faster execution: Matlab Parallel Computing Toolbox, IBM ILOG CPLEX. The default in the provided codes assumes the absence of the IBM Suite and the parallel computing toolbox.

    References :

    • Optimal treatment strategies in the context of ‘treatment for prevention’ against HIV-1 in resource-poor settings. S. Duwal, S. Winkelmann, C. Schütte and M. von Kleist, PLoS Comput. Biol., 11, e1004200, 2015 (open access)
    • Markov Control Processes with Rare State Observation: Theory and Application to Treatment Scheduling in HIV-1 S. Winkelmann, C. Schütte and M. von Kleist.  Communications in Mathematical Sciences 12, 859, 2014 (online)

    Transmic is a python program for clustering genetic sequence data based on a given phylogenetic tree structure. Its distinguishing feature is the capability to deal with various requirements imposed by epidemiological projects seeking to understand the properties of disease spread at the population level. Some of its most important features are:

    • a clustering metric based on a combination of branch distances with statistical confidence of the tree nodes (bootstrap or posterior probabilities)
    • patient-based clustering allowing for multiple follow-up sequences per individual as an option
    • support for different phylogenetic tree formats (newick / nexus)
    • the program allows for computation of multiple clusterings based on a range of clustering thresholds
    • visualization option for plotting key clustering properties

    The code and the program documentation can be found on github: https://github.com/kavehyousef/code. Further details of the underlying methodology are explained in the corresponding  publication.

    References :

    • Inferring HIV-1 transmission dynamics in Germany from recently transmitted viruses. K. Pouran Yousef, K. Meixenberger, M. R. Smith , S. Somogyi, S. Gromöller, D. Schmidt, B. Gunsenheimer-Bartmeyer. O. Hamouda, C. Kücherer and M. von Kleist , Journal of Acquired Immune Deficiency Syndromes, 73(3):356-363 (2016) (link)