Footprint based analysis and causal network contextualisation in SARS-CoV-2 infected A549 cell line

Introduction

This repository contains the code to reproduce the analysis presented in section mentioned in the project title contained in the article: A versatile and interoperable computational framework for the analysis and modelling of COVID-19 disease mechanisms

Procedure Details

We obtained the transcriptomics dataset from the GEO database with accession number GSE147507 Blanco-Melo et al., 2020. We extracted the series number 5 from the dataset, consisting of 2 conditions in triplicate, A549 cells treated with a mock and A549 infected with SARS-CoV-2, measured 24 hours after infection. Differential analysis of the transcript abundances was performed using DESeq2 Love et al., 2014. The resulting t-values of the differential analysis were used as input to estimate pathway activity deregulation using Progeny Schubert et al., 2018. The differential analysis t-values were also used to estimate the deregulation of TF activities using Dorothea Garcia-Alonso et al., 2019 as a source of TF-target regulon and the Viper algorithm Alvarez et al., 2016 to estimate the TF activity score.

Phosphoproteomic data of mock-treated and SARS-CoV.2 infected cells were extracted from Stukalov et al., 2021. Phosphosite differential analysis log2FC was used to estimate the deregulation of kinase activities using [https://doi.org/10.1101/822668] as a source of kinase-substrate interactions and a z-test to estimate kinase activity score [PMID: 32645325; PMID: 28200105].

Finally, we used Carnival [PMID: 31728204] with the COSMOS approach [PMID: 33502086] to connect the top 10 deregulated kinases with the top 30 deregulated TFs with a Prior Knowledge Network assembled from OmniPath resources [PMID: 33749993]. Progeny pathway activity scores were used to weigh the PKN and facilitate the optimal network search to connect kinases and TFs.