MRD ppmSeq Data
Last Update: February 5, 2024
Introduction:
The pursuit of rare variant detection takes on a role of paramount importance, particularly when examining Minimal Residual Disease (MRD) aiming to comprehend and monitor diseases at their most minuscule levels. Identifying these variants at extremely low allelic fraction not only informs clinicians about the efficacy of therapies but also serves as an early warning system, signaling the potential for disease recurrence.
This page shows the steps required to analyze the data in the MRD app note and generate the data showed in the figures presented on the app note.
This page goes through three main steps: (1) Analysis of Tumor/Variant data to generate mutational signature per patient (2) Analysis of cfDNA sample for standard and ppmSeq runs to generate a list of all candidate substations observed in the data and assign them calibrated quality scores and finally (3) calculate the overlap between the mutational signatures and the candidate variants in the plasma samples to estimate the tumor fraction in each cfDNA sample.
Data and code location:
File | Method | Average Size | Location |
---|---|---|---|
Input cfDNA cram | xGen | 384 GiB | s3://ultimagen-feb-2024-mrd/Application/MRD/cfDNA_xgen/XX.CRAM |
ppmseq | 123 GiB | s3://ultimagen-feb-2024-mrd/Application/MRD/cfDNA_ppmSeq/XX.CRAM | |
Input Tumor/Normal VCF | FFPE | 3.64 GiB | s3://ultimagen-feb-2024-mrd/Application/MRD/DV_signatures/XX_FFPE.ann.vcf.gz |
Fresh Frozen | 2.9 GiB | s3://ultimagen-feb-2024-mrd/Application/MRD/DV_signatures/XX_FreshFrozen.ann.vcf.gz | |
Intermediate Output cfDNA VCF (Featuremap) | xGen | 24 GiB | s3://ultimagen-feb-2024-mrd/Application/MRD/cfDNA_xgen/XX.with_ml_qual.vcf.gz |
ppmSeq | 9 GiB | s3://ultimagen-feb-2024-mrd/Application/MRD/cfDNA_ppmSeq/XX.with_ml_qual.vcf.gz | |
Final h5 file per cfDNA and tumor signature pair | xGen | 3.4 MiB | s3://ultimagen-feb-2024-mrd/Application/MRD/cfDNA_xgen/XX.tumor_fraction.h5 |
ppmSeq | 3.6 MiB | s3://ultimagen-feb-2024-mrd/Application/MRD/cfDNA_ppmSeq/XX.tumor_fraction.h5 |
How to:
*Somatic variant calling HowTo will be made available soon
WDLs:
*Somatic variant calling WDL will be made available soon
Prerequisite files and skills:
- Cram file generated on Ultima Genomics UG100 tool
- Familiarity with the Linux command line
- Familiarity with the SAM/BAM/CRAM format
Software and packages used:
- Capability to execute Workflow Description Language (WDL)
- Access to VariantCalling docker container
Running analysis pipelines:
Overview of processing steps
Step 1:
- Run Somatic Deep Variant wdl to generate mutational signatures for all patients with tumor/normal samples pairs
Step 2:
- Run Single read SNV.wdl to generate a featuremap VCF for all cfDNA samples
Step 3:
- Run MRDFeatureMap.wdlto generate the final overlapping