Pysam Fetch Behavior

Question

[rickyflintoff]

I wanted to fetch only certain columns from a tabix indexed file. Using Pysam's fetch command, I get the entire row in a string format. However, I would like to query only the exact column. Is there a quick way to accomplish this?

[Matt Shirley]

Could you be more specific? Do you want to fetch only certain columns from a subset of rows, or do you just want certain columns from the whole data?

[always_learning]

Use awk on pysam output for exact column.

[Matt Shirley]

If the OP is using pysam, it's a good guess that they are already in a Python environment. awk, while extremely useful, is a worse fit than simply accessing the AlignedRead object returned by the IteratorRow object returned by Samfile.fetch().

region = Samfile.fetch('chr1:1-1000')
positions = [read.pos for pos in region]

[rickyflintoff]

I am specifically working with TSV files. So, I here is an example of TSV file:

chr\tpos\tjudgement
1\t2389\tTrue
1\t2399\tTrue

When I use fetch to get row for chr 1 and pos 2399, I get the following: 1\t2399\tTrue However, I want a way to directly access judgement value for chr 1 and pos 2399. I'd like to directly call the exact column value. I cannot use awk. I was wondering if there is a direct way of doing so using Pysam. Thank you for all your help.

Answer 1

[Matt Shirley]

row = Tabixfile.fetch(reference='chr1', start=2399, end=2399, parser=asTuple)
judgement = row[2]

[always_learning]

import pysam
tabixfile = pysam.Tabixfile( "/usr/local/xyz.vcf" )

And now when I'm doing this:

row = tabixfile.fetch(reference='chr1', start=249240539, end=249240539)

but while doing

x = row[2]

I am getting this error.

Where I am going wrong actually? Any pointers?