Issue when converting from plink to eigenstrat using convertf

[jaurbanChicago]

Hello,

I have a vcf file and a plink file with 5,758,371 positions. The vcf was converted to plink using the plink software, and then, I used convertf to convert the plink (bed) into an eigenstrat file. After converting to eigenstrat, my eigenstrat file has 5,757,704 positions. Apparently, the convertf run had no issues since the .err file came out empty. Why would I be losing 667 positions? Thanks in advance

Best,
Jose Antonio

[bumblenick]

Does the convertf log file help? N

…

Sent from my iPhone

On Mar 6, 2023, at 11:13 PM, jaurbanChicago ***@***.***> wrote:  Hello, I have a vcf file and a plink file with 5,758,371 positions. The vcf was converted to plink using the plink software, and then, I used convertf to convert the plink (bed) into an eigenstrat file. After converting to eigenstrat, my eigenstrat file has 5,757,704 positions. Apparently, the convertf run had no issues since the .err file came out empty. Why would I be losing 667 positions? Thanks in advance Best, Jose Antonio — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.

[jaurbanChicago]

parameter file: plink2eigen_WGS_SriLankans.par
genotypename: /scratch/jaurban/SriLanka_HighCov_Genomes/vcf/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.bed
snpname: /scratch/jaurban/SriLanka_HighCov_Genomes/vcf/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.bim
indivname: /scratch/jaurban/SriLanka_HighCov_Genomes/vcf/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.fam
outputformat: EIGENSTRAT
genotypeoutname: /scratch/jaurban/SriLanka_HighCov_Genomes/eigenstrat/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.geno
snpoutname: /scratch/jaurban/SriLanka_HighCov_Genomes/eigenstrat/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.snp
indivoutname: /scratch/jaurban/SriLanka_HighCov_Genomes/eigenstrat/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.ind
familynames: NO
## convertf version: 5000
genetic distance set from physical distance
genotype file processed
numvalidind:   4884  maxmiss: 4884001
eigenstrat output
##end of convertf run


------------ Job WrapUp ------------

Job ID:            37825838.cri16sc001
User ID:           jaurban
Job Name:          plink2eigen_WGS_SriLankans.pbs
Queue Name:        mid
Working Directory: /scratch/jaurban/SriLanka_HighCov_Genomes/vcf
Resource List:     walltime=39:59:00,nodes=2:ppn=15,mem=32gb,neednodes=2:ppn=15
Resources Used:    cput=02:07:54,vmem=11390228kb,walltime=02:08:01,mem=10013956kb,energy_used=0
Exit Code:         0
Mother Superior:   cri16cn220

Execution Nodes: 
cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219

This is what the log file says!

Best,
JA

[bumblenick]

OK; can't help but possibilities are 1) convertf won't like snps without allele spec; 2) Only C1-24 is processed. I think you will have to do some debugging yourself. At least look at the missing snps for something funky Nick

…

On Mon, Mar 6, 2023 at 11:28 PM jaurbanChicago ***@***.***> wrote: parameter file: plink2eigen_WGS_SriLankans.par genotypename: /scratch/jaurban/SriLanka_HighCov_Genomes/vcf/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.bed snpname: /scratch/jaurban/SriLanka_HighCov_Genomes/vcf/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.bim indivname: /scratch/jaurban/SriLanka_HighCov_Genomes/vcf/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.fam outputformat: EIGENSTRAT genotypeoutname: /scratch/jaurban/SriLanka_HighCov_Genomes/eigenstrat/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.geno snpoutname: /scratch/jaurban/SriLanka_HighCov_Genomes/eigenstrat/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.snp indivoutname: /scratch/jaurban/SriLanka_HighCov_Genomes/eigenstrat/Panel_dbWGS_SriLankaModAnc.SNPbiallelic.rel.maf5p.maxSNPmiss10p.ind familynames: NO ## convertf version: 5000 genetic distance set from physical distance genotype file processed numvalidind: 4884 maxmiss: 4884001 eigenstrat output ##end of convertf run ------------ Job WrapUp ------------ Job ID: 37825838.cri16sc001 User ID: jaurban Job Name: plink2eigen_WGS_SriLankans.pbs Queue Name: mid Working Directory: /scratch/jaurban/SriLanka_HighCov_Genomes/vcf Resource List: walltime=39:59:00,nodes=2:ppn=15,mem=32gb,neednodes=2:ppn=15 Resources Used: cput=02:07:54,vmem=11390228kb,walltime=02:08:01,mem=10013956kb,energy_used=0 Exit Code: 0 Mother Superior: cri16cn220 Execution Nodes: cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn220 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 cri16cn219 This is what the log file says! Best, JA — Reply to this email directly, view it on GitHub <#86 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEE77B72D2LMHC6NVWNECZDW222QTANCNFSM6AAAAAAVR6XH24> . You are receiving this because you commented.Message ID: ***@***.***>

[janxkoci]

Another possibility are multiallelic SNPs - both Plink v1 and eigenstrat formats don't support those, while VCF has no problem with such positions.

Does your Plink file already show the reduction in number of sites? You can check with e.g. wc -l in.bim. On the VCF side, you can use something like bcftools view -H -m3 in.vcf | wc -l or even just awk '/#/ {next} $5 ~ /,/ {counter++} END {print counter}' in.vcf.