Problem converting STACKS .fa to migrate

[Diego Peralta]

Hi everybody,

I'm trying to convert populations.sample.fa from STACKS to migrate by using fasta2genotype.py and stacks2mig.py but both options give me error messages. My STACKS running gives me near 70k SNPs but I decided to reduce the number after reading several posts in this group. I could upload the first lines of my input files if you needed. 

So, for the first script, I used 5500 SNPs (file populations2.samples.fa) and 32 individuals (1 snp/locus and non-missing data among individuals), but after choosing the different options in the script, it stops with the next error:

/dmperalta/migrate_files/1.1snp# python2.7 fasta2genotype.py populations2.samples.fa migrate.whiteloci migrate.populations populations.snps.vcf infile1

Output type? [1] Migrate [2] Arlequin [3] DIYABC [4] LFMM [5] Phylip [6] G-Phocs [7] Treemix [8] Haplotype: 1

Loci to use? [1] Variable [2] All: 2

Coverage Cutoff (number reads for locus)? Use '0' to ignore coverage: 0

Filter for allele frequency? False alleles might bias data. [1] Yes [2] No: 2

Remove monomorphic loci? [1] Yes [2] No: 2

Filter for missing genotypes? These might bias data. [1] Yes [2] No: 2

Clip cut sites? These may bias data. [1] Yes [2] No: 2

Cataloging populations...

Cataloging loci...

Counting gene copies...

Counting locus lengths...

Traceback (most recent call last):

  File "fasta2genotype.py", line 620, in <module>

    num_sites = SeqSitesCount(sys.argv[1], cutsite1, cutsite2, checkbothends)

  File "fasta2genotype.py", line 594, in SeqSitesCount

    nextline = newfasta.next(); nextline = nextline.strip()

StopIteration

With the second script and using the the same fasta file, but with the entire title in the headers (file populations.sample.fa), I obtain the next error:

/dmperalta/migrate_stacks# cat populations.samples.fa | python2.7 stacks2mig.py

Traceback (most recent call last):

  File "stacks2mig.py", line 170, in <module>

    headers,sequences = read_stacks(infilename)

  File "stacks2mig.py", line 63, in read_stacks

    read_lines(sys.stdin,headers,sequences)

  File "stacks2mig.py", line 45, in read_lines

    index = h.index('>CLocus')

ValueError: '>CLocus' is not in list

Can someone help me with this? I was also trying to use the new vcf2migrate.py script, but it also gave me error messages

[Diego Peralta]

Hi everyone again, 

After one month of acquiring experience with the program and trying with different types of data files, I was able to convert my STACKS (2.59) fasta files to migrate using the converter fasta2genotype.py write by Paul Maier. 

I read several questions trying to convert fasta files to migrate so I thought to share my experience. It could be useful to say that I am new at bioinformatics so maybe someone could do that quickly and better. 

The converter needs a fasta and a vcf file from the same running, and a population file, which contains the number of locus assigned by gstacks or denovo_map.pl modules, the name of the sample and the name of the population, order as the popmap used in populations. I use the file populations.samples.fa produced by the flag --fasta-samples of populations and the populations.snps.vcf of the same running. 

First I converted the fasta file to the format of STACKS 1.12. The new fasta format has the names of the sample and chromosome in the header of each locus. I used this to eliminate them:

sed -i 's/_0 .*/_0/' populations.samples.fa

sed -i 's/_1 .*/_1/' populations.samples.fa

After that I checked the .fa file to search for the number assigned for my samples and use it to write the population file. The format needs to be like that:

SampleID        IndID   PopID

5       Q06B    A_Uru

1       PQ03    A_Uru

3       Q03     A_Uru

9       RN4     B_PatN

10      RN7     B_PatN

8       PV22    B_PatN

After this easy steps the files are ready for being convert to the migrate format. I hope that could be useful for someone :).

Best to all,

Diego 

[Quinn Carvey]

Hi Diego,

Just wanted to say thank you for posting your solution, it's useful years later!

Cheers,

Quinn