Test-drive the GATK tools and Best Practices pipelines on Terra
Check out this blog post to learn how you can get started with GATK and try out the pipelines in preconfigured workspaces (with a user-friendly interface!) without having to install anything.
Missing variants using the GATK best practices.
I am working with human whole exome (WES - Illumina, paired end) data and trying to perform variant calling by following the GATK best practices with GATK v188.8.131.52 installation(I know that there has been a new release few days back). GATK has been installed using conda config file as suggested on GATK installation page. Java version details are
openjdk version "1.8.0_192" OpenJDK Runtime Environment (Zulu 184.108.40.206-linux64) (build 1.8.0_192-b01) OpenJDK 64-Bit Server VM (Zulu 220.127.116.11-linux64) (build 25.192-b01, mixed mode)
I am comparing 2 variant datasets:
Variant calling done using GATK best practices taking Illumina Exome Paired end data from NovaSeq S2: Nextera Flex for Enrichment (12-plex, NA12878) with Twist Human Core Exome (Illumina BaseSpace public data)
VCF file downloaded from the same dataset from Illumina BaseSpace. Since the file (s55_NFE_Twist_NA12878-40M_S1.genome.vcf.gz) is genome VCF, it has been intersected with Twist exome BED file.
There are several SNV and INDELS which are present in DataSet#2 and not in DataSet#1 and we are trying to figure out why is that. For example, consider this region:
chr4:1388583 which has a
SNV (A>G/G) for the gene
CRIPAK present in DataSet#2 and not in DataSet#1.
What we have verified so far that:
--> This regions is well covered in DataSet#1 and DataSet#2.
--> That this region is present in the Twist Exome BED file that we are using
Could you provide us insights on how and what shall we do to compare the difference and figure out the exact reasons for the differences in the results. There are several such regions in both datasets.