Williamson, B. D. | Search bioRxiv

bioRxiv2020

Super LeArner Prediction of NAb Panels (SLAPNAP): A Containerized Tool for Predicting Combination Monoclonal Broadly Neutralizing Antibody Sensitivity

SummarySingle broadly neutralizing antibody (bnAb) regimens are currently being evaluated in randomized trials for prevention efficacy against HIV-1 infection. Subsequent trials will evaluate combination bnAb regimens (e.g., cocktails, multi-specific antibodies), which demonstrate higher potency and breadth in vitro compared to single bnAbs. Given the large number of potential regimens in the research pipeline, methods for down-selecting these regimens into efficacy trials are of great interest. To aid the down-selection process, we developed Super LeArner Prediction of NAb Panels (SLAPNAP), a software tool for training and evaluating machine learning models that predict in vitro neutralization resistance of HIV Envelope pseudoviruses to a given single or combination bnAb regimen, based on Envelope amino acid sequence features. SLAPNAP also provides measures of variable importance of sequence features. These results can rank bnAb regimens by their potential prevention efficacy and aid assessments of how prevention efficacy depends on sequence features. Availability and ImplementationSLAPNAP is a freely available docker image that can be downloaded from DockerHub (https://hub.docker.com/r/slapnap/slapnap). Source code and documentation are available at GitHub (respectively, https://github.com/benkeser/slapnap and https://benkeser.github.io/slapnap/). ContactDavid Benkeser, benkeser@emory.edu

bioinformatics↗

bioRxiv2019

A multi-view model for relative and absolute microbial abundances

The absolute abundance of bacterial taxa in human host-associated environments play a critical role in reproductive and gastrointestinal health. However, obtaining the absolute abundance of many bacterial species is typically prohibitively expensive. In contrast, relative abundance data for many species is comparatively cheap and easy to collect (e.g., with universal primers for the 16S rRNA gene). In this paper, we propose a method to jointly model relative abundance data for many taxa and absolute abundance data for a subset of taxa. Our method provides point and interval estimates for the absolute abundance of all taxa. Crucially, our proposal accounts for differences in the efficiency of taxon detection in the relative and absolute abundance data. We show that modeling taxon-specific efficiencies substantially reduces the estimation error for absolute abundance, and controls the coverage of interval estimators. We demonstrate the performance of our proposed method via a simulation study, a sensitivity study where we jackknife the taxa with observed absolute abundances, and a study of women with bacterial vaginosis.

genomics↗