Different Chromosome

[jiangweixia]

Hi Juliana, Thank you for sharing your genetic QC pipeline—it has been incredibly helpful. I have a question regarding whole-genome sequencing (WGS) data. Due to its large size, it's often to separate the data by chromosomes. I noticed that your input data is a single, combined vcf.gz file for all samples.
Given that I have data spanning all 24 chromosomes, what would you recommend as the best approach to handle this? Should I process the data as a whole (very large if combined), or is there an efficient way to manage it by chromosome when I use your code? Thank you again for your assistance.

Best regards,
Xia

[acostauribe]

Hi Xia,

You are right. Some pipelines split the data into chromosomes to make it easier/faster if you have small computational power or a large dataset. That approach has some advantages and disadvantages. For example, there are some metrics for which you could split the data into chromosomes and analyze each one independently, but there are some metrics like heterozygosity or kinship that need you to use the entire genome.
At the end of the day it is all about 'which is the question you are asking'?
If you care about genotype depth (DP) and genotype quality (GQ) you could split the data into chromosomes before doing any of the genotype quality filters, since these are the processes that take the longest. If you do not care about these, you can import the data into plink and analyze genomes as a whole.

I hope this helps!

J