snpgenie\_within\_group.pl: Question about input file formats

[annasimonsen]

I have several thousand fasta files, each fasta file represents a single gene containing all sequenced individuals from a single population. Each fasta file is a nucleotide alignment, which I have attempted to framecode align using TranslatorX. I would like to calculate piN/piS for each gene using snpgenie_within_group.pl

I do not have a gtf file. Can I run snpgenie_within_group.pl without the gtf file? If not, can you offer any guidance on how would I format this type of sequence data and generate the correctly formatted gtf file?

Thanks

[singing-scientist]

Thanks a lot for the question @annasimonsen! Unfortunately I wrote the script with chromosomes (not genes) in mind and did not have enough foresight to allow flexible usage without the GTF. However, I think your pipeline could auto-generate temporary GTF files one the fly. For example, suppose your directory contained three files: gene1.fasta, gene2.fasta, and gene3.fasta. You'll probably be looping through these files somehow, perhaps using a wrapper Unix script. When you hit gene1.fasta, you can determine the length of the sequences inside — I think bioawk has something ready made, or you could get clever with cat, grep, and awk — and then write a file called gene1.gtf. For example, if gene1.fasta is an alignment of sequences with 693 nucleotides, you'd simply write the 1-line GTF file:

my_temp_gtf my_temp_gtf CDS 1 693 . + 0 gene_id "gene1";

Then provide that temp file as an argument to SNPGenie, and delete when finished (or whatever you prefer). In other words, ever temp GTF file you produce will be a single line that species one gene beginning at 1 (in every case) and ending at the last site (i.e. length).

Let me know if that helps!

Chase

[annasimonsen]

Thanks for your quick reply! Ill let you know how that goes.