Tan, K. P. | bioRxiv Science

bioRxiv2021

Data-driven analysis of COVID-19 reveals specific severity patterns distinct from the temporal immune response

Key immune signatures of SARS-CoV-2 infection may associate with either adverse immune reactions (severity) or simply an ongoing anti-viral response (temporality); how immune signatures contribute to severe manifestations and/or temporal progression of disease and whether longer disease duration correlates with severity remain unknown. Patient blood was comprehensively immunophenotyped via mass cytometry and multiplex cytokine arrays, leading to the identification of 327 basic subsets that were further stratified into more than 5000 immunotypes and correlated with 28 plasma cytokines. Low-density neutrophil abundance was closely correlated with hepatocyte growth factor levels, which in turn correlated with disease severity. Deep analysis also revealed additional players, namely conventional type 2 dendritic cells, natural killer T cells, plasmablasts and CD16+ monocytes, that can influence COVID-19 severity independent of temporal progression. Herein, we provide interactive network analysis and data visualization tools to facilitate data mining and hypothesis generation for elucidating COVID-19 pathogenesis.

immunology↗

bioRxiv2021

Packpred: Predicting the functional effect of missense mutations

1.Predicting the functional consequences of single point mutations has relevance to protein function annotation and to clinical analysis/diagnosis. We developed and tested Packpred that makes use of a multi-body clique statistical potential in combination with a depth dependent amino acid substitution matrix (FADHM) and positional Shannon Entropy to predict the functional consequences of point mutations in proteins. Parameters were trained over a saturation mutagenesis data set of T4-lysozyme (1966 mutations). The method was tested over another saturation mutagenesis data set (CcdB; 1534 mutations) and the Missense3D data set (4099 mutations). The performance of Packpred was compared against those of six other contemporary methods. With MCC values of 0.42, 0.47 and 0.36 on the training and testing data sets respectively, Packpred outperforms all method in all data sets, with the exception of marginally underperforming to FADHM in the CcdB data set. On analyzing the results, we could build meta servers that chose best performing methods of wild type amino acids and for wild type-mutant amino acid pairs. This lead to an increase of MCC value of 0.40 and 0.51 for the two meta predictors respectively on the Missense3D data set. We conjecture that it is possible to improve accuracy with better meta predictors as among the 7 methods compared, at the least one method or another is able to correctly predict [~]99% of the data.

bioinformatics↗