Sequencing tools | Check sample missmatches

[efigb]

Usage of the Picard (GATK) tool CrosscheckFingerprints that uses an haplotype database to identify sample missmatches between case-normal sample pairs.

[efigb]

adapt instructions:

INSTRUCTIONS

1/ create a conda environment with picard installed via anaconda:

conda install bioconda::picard

2/ download the haplotype map database from picard from broad institute google cloud: https://console.cloud.google.com/marketplace/product/broad-institute/references

the haplotype_map file .txt file is: hg38 > v0 > Homo_sapiens_assembly38.haplotype_database.txt
copy the link and in terminal

wget link

3/ assure the read groups are annotated in the normal-paired samples to compare with samtools. use the samtools conda environment to do that:

samtools view -H sample_recalibrated.bam | grep "^@rg"

This command will display all the @rg (read group) tags present in the BAM file. A typical read group (@rg) line might look like this:

@rg ID:group1 SM:sample1 PL:ILLUMINA LB:lib1 PU:unit1

normal_paired samples should show different SM field in @rg. We will use this information in SM with picard to distinguish normal-pairs:

4/ Activate conda picard environment to run the command:

picard CrosscheckFingerprints -INPUT /workspace/nobackup2/melups/xpc_patients/20240822_set1_sarek/preprocessing/recalibrated/SA002_N1/SA002_N1.recal.bam -INPUT /workspace/nobackup2/melups/xpc_patients/20240822_set1_sarek/preprocessing/recalibrated/SA002_T2/SA002_T2.recal.bam -HAPLOTYPE_MAP ./Homo_sapiens_assembly38.haplotype_database.txt -LOD_THRESHOLD -5 -EXPECT_ALL_GROUPS_TO_MATCH true -CROSSCHECK_BY SAMPLE -MATRIX_OUTPUT SA002.matrix.txt

-INPUT is the bam (indexed in the directory) of the normal/tumor sample for the same patient
-HAPLOTYPE_MAP is the haplotype map database downloaded from broad institute files with haplotye regions to check
-LOD_THRESHOLD this is the default
-CROSSCHECK_BY SAMPLE very important this will assure that samples are distinguished based on SM field in readgroup field of bam files
-MATRIX_OUTPUT is easier to interpret the result in a matrix

Interpretation

The tool uses the haplotype database (Homo_sapiens_assembly38.haplotype_database.txt) to find any mismatches between samples and calculates a LOD score and also tumor-aware LOD score which can help assess identity even in the presence of a severe LOH sample with high purity.

The scale is logarithmic (base 10), so a LOD of 6 indicates that it is 1,000,000 more likely that the data matches the genotypes than not. A negative value indicates that the data do not match. A score that is near zero is inconclusive and can result from low coverage or non-informative genotypes.

When there is a sample mismatch it displays negative value of LOD and in terminal appears the message of the missmatch between samples.

Examples in:

/data/bbg/nobackup2/melups/xpc_patients/set1/20240913_set1_check_sample_match

TO DO:

The downloaded haplotype database should be moved to datasets with a readme file.