COVID-19 drug repurposing

COVID-19 drug repurposing identifies and evaluates potential therapeutic targets and candidate drugs for COVID-19 by integrating in silico immune system protein interactome analysis, single-cell RNA sequencing data from human tissues, and deep neural network–based drug activity prediction.

Key Features:

• Immune System Protein Interactome Network Analysis: Performs in silico analysis of high-confidence immune system proteins within cells co-expressing ACE2 and TMPRSS2 to identify proteins overexpressed in nasal goblet secretory cells, lung type II pneumocytes, and ileal absorptive enterocytes.
• Single-Cell RNA Sequencing: Integrates single-cell RNA-seq data from human tissues to map cell-type-specific gene expression patterns relevant to COVID-19 pathology.
• Artificial Neural Networks: Employs fully connected multitask deep neural networks to predict activity for 10,672 drugs, including approved, under investigation, and experimental compounds, with performance assessed by area under the receiver operating characteristic curve (AUC).

Scientific Applications:

• Drug Repurposing: Identifies existing drugs with predicted activity against targets implicated in COVID-19 for prioritization in further experimental or clinical evaluation.
• Therapeutic Target Identification: Pinpoints overexpressed immune-related proteins in specific cell types that may serve as mechanistic targets for intervention.
• Predictive Modeling: Provides in silico efficacy predictions to inform selection of candidate compounds and design of downstream validation studies.

Methodology:

In silico screening of immune system proteins in cells co-expressing ACE2 and TMPRSS2, integration with single-cell RNA-seq expression data from human tissues, and multitask fully connected deep neural network prediction of drug activity for 10,672 compounds with AUC-based evaluation.