We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

# STR annotation

Member

May I know how is this defined?

## Best Answer

• Broad Institute ✭✭
Accepted Answer

If STR is present in the INFO field, then indel is a tandem repeat. If present, then annotation RU gives a string representing the bases that form the repeat unit, and RPA are the repeats per allele.
For example, if a (CA)^3 becomes (CA)^4 (2 bp insertion), then STR is present, RU = CA, and RPA = 3,4.

## Answers

• Broad InstituteMember ✭✭
Accepted Answer

If STR is present in the INFO field, then indel is a tandem repeat. If present, then annotation RU gives a string representing the bases that form the repeat unit, and RPA are the repeats per allele.
For example, if a (CA)^3 becomes (CA)^4 (2 bp insertion), then STR is present, RU = CA, and RPA = 3,4.

• Member

Dear @delangel,

I wonder then, if I get a lot of negative values, is this because of STRs that are deletions to the reference genome?

Best

• Member

How was the STR flagged in the INFO column? Please see the igvshot, and the mutect2 tumoronly
output here.

```chrX    15349851    .   TA  T   .   PASS    CONTQ=93;DP=26;ECNT=1;GERMQ=18;MBQ=35,36;MFRL=256,222;MMQ=60,60;MPOS=40;POPAF=7.30;RPA=2,1;RU=A;SAAF=0.283,0.273,0.308;SAPP=0.017,0.026,0.957;STR;TLOD=22.22    GT:AD:AF:DP:F1R2:F2R1:OBAM:OBAMRC   0/1:18,8:0.321:26:9,3:9,5:false:false
```

Thank you.

• Member
edited June 2019

Any GATKers can give me some suggestions ?
Thank you very much.

• Cambridge MAMember, Administrator, Broadie, Moderator admin

Hi @xiucz

Apologies for the delay in replying. In order to clarify, is your question "why the tool flagged this an STR"? Could you also provide a bam and bamout screenshots of this region to compare?

• Cambridge MAMember, Administrator, Broadie, Moderator admin

Also please post the version of Mutect2 used and the exact command.

• Member
edited June 2019

Thank you for your reply.

Yes, my question is "why the tool flagged here an STR"? It seems that there is only one repeated allele.
Screenshot and commands:

```~/gatk-4.1.0.0/gatk Mutect2 \
-R ~/database/hg19/gatk_bundle/ucsc.hg19.fasta \
-I ~/sample.recalibrated.bam -tumor samplename \
-L ~/exon.bed --interval-padding 250 \
--germline-resource ~/af-only-gnomad.raw.sites.hg19.vcf.gz \
--tmp-dir ~/tmpdir -O sample.vcf \
-bamout sample.recalibrated.mt2.bam

~/gatk-4.1.0.0/gatk GetPileupSummaries \
-I ~/sample.recalibrated.bam \
-L ~/exon.bed --interval-padding 250 \
-V ~/small_exac_common_3_hg19.vcf.gz \
-O sample.tumor.getpileupsummaries.table

~/gatktools/gatk-4.1.0.0/gatk CalculateContamination \
-I sample.tumor.getpileupsummaries.table \
-O sample.calculatecontamination.table

~/gatk-4.1.0.0/gatk FilterMutectCalls \
-V sample.vcf \
--contamination-table sample.calculatecontamination.table \
-O sampple.mutect2_oncefilter.vcf
```

Xiucz

• Cambridge MAMember, Administrator, Broadie, Moderator admin

Hi @xiucz

This is what the Mutect2 dev said:
Only the "slippage" filter (previously "str_contraction"), not the STR INFO, means that M2 rejects a variant. The STR INFO field applies for any indel of one or more units of a repeat, even if, as in this case, there are as few as two repeat units in the reference. M2 would never filter such a short "STR," however. Currently we don't filter for STRs smaller than 8 reference bases, and we don't necessarily filter larger ones. We are about to come out (within two months) with a much-improved polymerase slippage filter.

Sign In or Register to comment.