COVID-19 exome project

Directory feature_processing

Preparation of datasets, exploratory data analysis, filtering, normalization of data.

• feature_mapping.py, extra_feature_mapping.py - name mappings of features (from Russian to English short abbreviations).
• part1_ids_preprocessing.ipynb - preprocessing of VCF and phenotypes ids.
• part2_phenotype_preprocessing.ipynb - cleaning and preparation of phenotypes, renaming of cols, etc.
• part2.5_new_table_analysis.ipynb - the same, but with extra-features.
• part3_march_pheno_eda_and_normalysing.ipynb - EDA, filtering, normalization of data.

Directory cvas_and_rvas

Common variants association study (CVAS) and Rare variants association study (RVAS).

• sd3_gwas_com.ipynb - CVAS.
• final_rwas_pipeline_p1_hail_prepare.ipynb - first step of RVAS (preparation of tables).
• final_rwas_pipeline_p2_statistics_and_plots.ipynb - second step of RVAS (tests and plots).
• out_hail_gwas_com_sd3/ - directpry with p values of cvas.
• out_hail_rvas/ - directory with p values of rvas.

Directory risk_score_and_regression

Validation of found SNPs: its annotation and statistical checks.

• Variants_annotations_and_score_calculation.ipynb - the main script with all annotations, statistics counts, etc (columns_to_check.py need for this notebook).
• draw_data/ - directory with datasets for R plots.
• other_data/ - directory with outputs of this script, which are not needed for drawing pictures.

Directory r_draw

R code for drawing figures, and figures itself.

• 1_MH_QQ.R - draw Manhattan and QQ plots (files: Rectangular-Manhattan.*t*.pdf and QQplot*.pdf respectively).
• 2_PCA.R - draw plots for principal components from EDA (for sex and death). Images: images/pca_*.pdf.
• 3_violin_regression.R - draw violin plots and regression on SNPS and associated features (from data/regression/regr_rs*_*.tsv). Images: images/regr_<rs>_<feature>.pdf.
• 4_boxplots.R - draw boxplots for features by death and severity (from data/boxplots_analyses.tsv). Images: images/bozplot_violin_<death/severity>_<feature>.pdf.
• 5_histplots.R - draw histplots. Imaged data:
  • images/histogram_<feature>___top_10_score.pdf - histograms for score by severity/death/storm;
  • images/score_hist.pdf - histogram pf the snps' score;
  • images/histogram_<death/severity>_<feature>.pdf - histograms of features by death/severity.
• images/ - drawn figures.