I'm sorry, could you suggest, why after genotype calling of a haploid
organism
with SingleSampleVariantsDetector with -ploidy 1 option the resultant .vcf
contains '0/1' genotypes, please?
I've used the following command:
ngsep SingleSampleVariantsDetector -runRep -runRD -runRP -minQuality 30
-ploidy 1 -sampleId {some sample Id} -r {reference} -i {input} -o {some
sample}_Q30
Such genotypes are also present in Demo_ann_q40_s_fi_I2_noREP_noCNV.vcf.
I've done calling on the same bam file with another caller, and it tends to
render the genotypes at the same positions as '1' in the vcf, but I want to
use ngsep.
I've done genotype calling from the same bam file using another caller, and
it tends to render genotypes at the corresponding positions as '1', but I
want to use ngsep.
Please, help me.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
Anonymous
-
2020-11-20
I have a problem with heterozygous calls in the .vcf for a haploid organism, particularly I get about 20% 0/1 calls, even using -ploidy 1 option. Ran SingleSampleVariantsDetector and bcftools genotyping on one bam from bowtie alignment, and ngsep genotyped many positions as 0/1, where bcftools produced haploid calls. Could you tell me, what could be the reson, please? When does ngsep produce 0/1 calls even with -ploidy 1 option?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks for your interest in NGSEP. For haploid samples what we do by default is to reduce the prior heterozygosity rate to 10E-6 to still allow the possibility of heterozygous calls. Although in principle this probability should be set to zero for haploid samples, keeping a low probability has been useful in our experience (for example in yeast) to aid with identification of local duplications and even aneuploidies. You can term these calls as heterogeneous instead of heterozygous but in any case I think it is useful to see them and to analyze the regions where these calls are located. This is also helpful in some cases to validate if your sample is actually haploid.
If you definitely think that your sample should not have any heterozygous call, you can just set the prior heterozygosity rate (option -h ) to zero (that is probably what bcftools is doing).
Let me know how things go
Jorge
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
Anonymous
-
2020-12-12
Thank you so much! The sample strain was additionally tested in the laboratory and it appeared to be triploid! I called variants again with ploidy -3 option, but in the .vcf there are still mostly diploid calls. But anyway, ngsep helped to uncover that the strain is triploid!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
Anonymous
-
2020-12-12
Kind regards,
Dasha
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I am glad to hear that NGSEP helped to find out the real ploidy of the sample. Regarding encoding of triploid calls, some time ago we decided to use an alternative field for the relative alleles in a polyploid call called ACN instead of using the GT field. For example, if you have a heterozygous genotype call in a triploid sample for a biallelic SNV, the ACN field should have either "1,2" or "2,1" depending on the number of times the reference and the alternative alleles are present.
Best regards
Jorge
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm sorry, could you suggest, why after genotype calling of a haploid
organism
with SingleSampleVariantsDetector with -ploidy 1 option the resultant .vcf
contains '0/1' genotypes, please?
I've used the following command:
ngsep SingleSampleVariantsDetector -runRep -runRD -runRP -minQuality 30
-ploidy 1 -sampleId {some sample Id} -r {reference} -i {input} -o {some
sample}_Q30
Such genotypes are also present in Demo_ann_q40_s_fi_I2_noREP_noCNV.vcf.
I've done calling on the same bam file with another caller, and it tends to
render the genotypes at the same positions as '1' in the vcf, but I want to
use ngsep.
I've done genotype calling from the same bam file using another caller, and
it tends to render genotypes at the corresponding positions as '1', but I
want to use ngsep.
Please, help me.
I have a problem with heterozygous calls in the .vcf for a haploid organism, particularly I get about 20% 0/1 calls, even using -ploidy 1 option. Ran SingleSampleVariantsDetector and bcftools genotyping on one bam from bowtie alignment, and ngsep genotyped many positions as 0/1, where bcftools produced haploid calls. Could you tell me, what could be the reson, please? When does ngsep produce 0/1 calls even with -ploidy 1 option?
Hi
Thanks for your interest in NGSEP. For haploid samples what we do by default is to reduce the prior heterozygosity rate to 10E-6 to still allow the possibility of heterozygous calls. Although in principle this probability should be set to zero for haploid samples, keeping a low probability has been useful in our experience (for example in yeast) to aid with identification of local duplications and even aneuploidies. You can term these calls as heterogeneous instead of heterozygous but in any case I think it is useful to see them and to analyze the regions where these calls are located. This is also helpful in some cases to validate if your sample is actually haploid.
If you definitely think that your sample should not have any heterozygous call, you can just set the prior heterozygosity rate (option -h ) to zero (that is probably what bcftools is doing).
Let me know how things go
Jorge
Thank you so much! The sample strain was additionally tested in the laboratory and it appeared to be triploid! I called variants again with ploidy -3 option, but in the .vcf there are still mostly diploid calls. But anyway, ngsep helped to uncover that the strain is triploid!
Kind regards,
Dasha
Hi Dasha
I am glad to hear that NGSEP helped to find out the real ploidy of the sample. Regarding encoding of triploid calls, some time ago we decided to use an alternative field for the relative alleles in a polyploid call called ACN instead of using the GT field. For example, if you have a heterozygous genotype call in a triploid sample for a biallelic SNV, the ACN field should have either "1,2" or "2,1" depending on the number of times the reference and the alternative alleles are present.
Best regards
Jorge