GRO-seq data analysis

[huo]

Hi,

I need to analyze publicly available gro-seq data. But I am new to these kind of analysis.

1) It is written in the GEO
"Bed files are tab-separated text files in which columns represent: chrom, chromStart (5' End of the read), chromEnd (chromStart+1), name (unused always 'n'), score (the number of mismatches), and strand. "

I need to find tags near TSS of genes. For genes on the minus strand what should be my approach?

2) I am using refseq annotation files from USCD. For example for gene BLNK (NM_001114094) which is on negative strand, txstart (97941444) and txend (98021323)

From below post

https://lists.soe.ucsc.edu/pipermail/genome/2009-June/019359.html

"Genomic coordinates that we store and display are always relative to the
positive strand of the reference genome. We reverse-complement genomic
sequence when a gene is on the reverse strand, but genomic coordinates
are still forward. So for a gene on the reverse strand, txStart is its
3' end, txEnd is its 5' end, and the exon coordinates are listed 3' to
5' (backwards)."



Therefore I am treating txend (98021323) is txstart and txstart (97941444) as txend.



3) When I try to find corresponding gro-seq tags what should be my approach for negative strand

For example a chr is 1000 bp. a gene in negative strand according to ucsd convention id from 10(txend) to 100(txstart).

According to gro-seq files since it is starting from 5' coordinates that I should look

a) 10 (txend) - 100 (txstart)

b) 900(txend) - 990(txstart)

I am a bit confused. Can anyone help me?