raimondilab Goto Github PK

an-explainable-model-of-host-genetic-interactions-linked-to-covid19-severity

This project focused on the mapping of the host-genetics factors determining COVID-19 severity using Machine learning approaches (supervised, unsupervised Machine learning methods, Pathway signaling processes, and Open Targets web-based Platform). Our study utilized the whole-exome sequencing genome dataset of 2000 European descent patients collected from the GEN-COVID Multicenter Study group (https://clinicaltrials.gov/ct2/show/NCT04549831) coordinated by the University of Siena. The whole-exome genome sequencing dataset contained 1.057M genetic variants of the patients. We used the 2000 patients’ original phenotype information to filter only patients with severity and asymptomatic across all classification criteria (841 patients). We introduced an innovative variant screening strategy that applied K-stratified fold splits of the original dataset to randomly draw a unique 5-fold pool of variants using the patients’ original phenotype information (841 unique patients).

cellhit

Code to replicate "Learning and actioning general principles of cancer cell drug sensitivity"

covid-19-severity-host-genetic-predictor-model-explanation

expansion

EXploring Protein AlterNative SplIcing cONsequence

gpcr_structure_analysis

gpcrsignalingaxes

Explore the dysregulation of GPCR ligand signaling systems in cancer genomics datasets.

precogx

A predictor of GPCR couplings with G-proteins/B-arrs using Transformers