Primary & Secondary analyses pipeline
Version 1
Genomic raw data are processing using Dragen hardware.
| Software version | 4.2.4 |
| Reference genome | Illumina DRAGEN Multigenome Graph Reference hg38 (alt-masked_graph+cnv+hla+rna_v3) |
All callers from DNA pipeline have been activated (Small variants, SV, CNV, HLA, PGx, SMN, SMA, …).
The CRAM format is selected as the output for alignment (lossless, V3.0). For the variants, we output both gVCF & VCF files.
Pre-Aggregation QC pipeline
Version 1
This pipeline tests each WGS sample, independently. If a sample fails this test, then it is excluded from the aggregation and not released or reported in this document. These tests detect sample swaps, cross-individual contamination, and sample preparation/sequencing errors. The list of specific QC processes is listed in the table below:
QC procedure | Criteria |
Sequencing metrics | for sample sequenced before Oct 2023: Autosomal mean depth of coverage >= 15 X for samples sequenced after Oct 2023:
Median insert size : >= 250 |
Aggregation Pipeline
Version 1
Samples that PASS the pre-aggregation QC pipeline (per sample QC) are aggregated into one multi-sample VCF file. Sample are aggregated using Dragen Iterative gVCF genotyper. To accelerate the aggregation and the downstream analyses (parallel processing), the genome is split into 100 shards.
The output of this pipeline is a multi-sample VCF. One multi-sample VCF is generated per shard in addition to the global multi-sample VCF files (all shards concatenated).