xQTLs

Scripts for various molecular quantitative trait locus (xQTL) analysis.

All Perl, Python, R, and Bash Shell scripts were used to eQTLs and RNA-m6A QTLs analysis by integrating QTLtools and MASH.

Steps:

1. Quantify reads density from ChIP-seq or MeRIP-seq data by 1a_QTLtools_quan.peaks.pl, or quantify gene expression level from RNA-seq data by 1b_QTLtools_quan.gene.pl. QTLtools quan is invoked in line 125, GTF file may need to be changed based on your file name. More details please see perl 1a_QTLtools_quan.peaks.pl -help or perl 1b_QTLtools_quan.gene.pl -help.

2. Get value log2((FPKM_IP+1)/(FPKM_INPUT+1)) of each peak for each MeRIP-seq sample by 2a_log2ratio.peaks.pl. And get value log2(TPM_INPUT+1) of each gene for each RNA-seq sample (Input of MeRIP-seq) by 2b_log2TPM.gene.pl. More details please see perl xxx.pl -help.

3. Merge all files from each sample into one file to satisfy the input format of QTLtools cis and QTLtools trans by 3a_merge.peaks.pl or 3b_merge.gene.pl. Please ensure that the colomn orders of the GFT file and the merged file are the same. More details please see perl xxx.pl -help.

4. Further filter m6A-epaks or genes by 4a_keptRegions.peaks.R or 4b_keptRegions.gene.R. We can remove peaks or genes with low reads desnity or low variance. More details please see code of 4a_keptRegions.peaks.R or 4b_keptRegions.gene.R

5. PCA for genotype by 5_QTLtools_pca.SNP.sh. PCA and PEER factor analysis for phenotype please see the codes under folder covariates.

6-8. Identify cis- or trans-QTLs by invoking QTLtools cis or QTLtools trans.