Compare Tum. only vs co-culture NK-cells.
Check if there are any mutations arising during the co-culture process.
Here is the email:
I would like to ask you, if you could help me with another analysis using my RNAseq data from the in vitro experiment, in which we co-cultured mouse tumor cells with NK cells. Dagmar suggested to do a variant calling, to see if we would be able to detect mutations/SNPs like this. If we find anything, we could subtract the mutations from the tumor only samples from the ones in the tumor + NK cell samples. Like this we see, which mutations arise during the co-culture. Dagmar already has some experience with this and sent me how she does it (email below).
Hi Michelle,
Here some information how I did the variant calling using RNA-seq data:
I used a script adapted according to the Genome Analysis Toolkit (GATK) Best Practices recommendations.
This is the newest version for hg38. I “-ERC NONE” in this case, otherwise it generates huge files.
I was only interested in certain regions, that’s why there is a samtools view step in the beginning.
~/bioinf_isilon/Research/STREHL/zStrehl_Halbritter/projects/AMKL/variant_calling_CBFA2T3GLIS2_AMKL/GATK_RNAseq/GATK_variant_calling_HaplotypeCallerStd_hg38_V1.sh
This is the old version for hg19 creating more control plots. However, I never checked them and therefore skipped these steps.
The hg38 script uses also a newer GATK-toolbox version.
~/bioinf_isilon/Research/STREHL/Internal/Dagmar/projects/Leukemia_pipeline/pipeline_leukemia_GATK_hg19_V6_20230517.sh
If I remember correctly, you need the STAR-aligned file as input, the reference fasta file and a dbsnp file (for recalibration).
I once saved this links in my comments to variant calling files à might be helpful:
genomics/workflows/gatk-mouse-mm10.md at master · igordot/genomics (github.com)GATK for inbred mouse - My Computational Genomic Playground (zqfang.github.io)
In the end I would intersect the unfiltered vcf-files from the NK cell resistant tumor to the “normal” tumor using bcftools. I would expect a rather short list of variants, should be quick using the online tool for annotation.
Variant Effect Predictor - Mus_musculus - Ensembl genome browser 111
If you need more or better annotation, you might lift it using crossmap and do the standard VEP annotation.
Hope that’s helpful.
Cheers, Dagmar
Compare Tum. only vs co-culture NK-cells. Check if there are any mutations arising during the co-culture process.
Here is the email: