Aligning WGS to PHG: keyfile

Question

[kathrynm]

Hello,

I am aligning several sets of paired end reads (ie two to four SRR files each separated into _R1 and _R2) per taxa. Do all of the files go into a single comma separated list in the keyfile (so 4-8 fastq files each)? Or do I need a separate entry for each SRR file?

Thanks, Kathryn

Answer 1

[zrm22]

Hi Kathryn,

Are you creating PHG Haplotypes using the WGS or are you Path finding using an existing PHG with WGS being the samples to genotype?

If you are creating PHG Haplotypes(Using the CreateHaplotypesFromFastq.groovy script), and they are paired end you will need to have a single line in the keyfile for each pair with them being comma separated. If you have 4 files, you will need to have 2 entries in the key file.

If you are using an existing PHG and running Path finding, you will still need to have a key file record for each pair, but it uses a different keyfile format. This wiki page(https://bitbucket.org/bucklerlab/practicalhaplotypegraph/wiki/UserInstructions/ImputeWithPHG_findPathKeyFiles) will show the format you need to use. The main difference is that there is a filename column and a filename2 column you will need to use.

[kathrynm]

I am adding haplotypes to the db, so I would use the same taxa name for both+ entries, right? Does that mean I can also add WGS for a taxa that already has an assembly loaded? Have you found value in using both assemblies and WGS to make haplotypes?