Skip to content

Latest commit

 

History

History
128 lines (98 loc) · 5.23 KB

deepsomatic-case-study-wgs-tumor-only.md

File metadata and controls

128 lines (98 loc) · 5.23 KB
Raw

DeepSomatic WGS tumor-only case study

In this case study, we show an example of running DeepSomatic on WGS tumor-only data. We use HCC1395 as an example for this case study.

Data details

For this case-study, we use HCC1395 as an example. We run the analysis on chr1 that we hold out during training.

Please see the metrics page for details on runtime and data.

Allele frequency channel

For accurate tumor-only calling, we use the allele-frequency channel that uses 1000 genomes variant calls using DeepVariant to filter out germline variants during inference. Currently, the default VCF is set to variant calls against GRCh38 reference. If you want to customize this to your VCF then please do so by using --population_vcfs parameter.

Prepare environment

Tools

Docker will be used to run DeepSomatic and hap.py,

Download input data

We will be using GRCh38 for this case study.

BASE="${HOME}/deepsomatic-wgs-tumor-only-case-study"

# Set up input and output directory data
INPUT_DIR="${BASE}/input/data"
OUTPUT_DIR="${BASE}/output"

## Create local directory structure
mkdir -p "${INPUT_DIR}"
mkdir -p "${OUTPUT_DIR}"
mkdir -p "${OUTPUT_DIR}/sompy_output"

# Download bam files to input directory
HTTPDIR=https://storage.googleapis.com/deepvariant/deepsomatic-case-studies/deepsomatic-chr1-case-studies
# Download the reference files
curl ${HTTPDIR}/GCA_000001405.15_GRCh38_no_alt_analysis_set.chr1.fna > ${INPUT_DIR}/GCA_000001405.15_GRCh38_no_alt_analysis_set.chr1.fna
curl ${HTTPDIR}/GCA_000001405.15_GRCh38_no_alt_analysis_set.chr1.fna.fai > ${INPUT_DIR}/GCA_000001405.15_GRCh38_no_alt_analysis_set.chr1.fna.fai

# Download the bam file
curl ${HTTPDIR}/HCC1395_wgs.tumor.chr1.bam > ${INPUT_DIR}/HCC1395_wgs.tumor.chr1.bam
curl ${HTTPDIR}/HCC1395_wgs.tumor.chr1.bam.bai > ${INPUT_DIR}/HCC1395_wgs.tumor.chr1.bam.bai

# Download truth VCF
DATA_HTTP_DIR=https://storage.googleapis.com/deepvariant/deepsomatic-case-studies/SEQC2-S1395-truth
wget -P ${INPUT_DIR} "${DATA_HTTP_DIR}"/High-Confidence_Regions_v1.2.bed
wget -P ${INPUT_DIR} "${DATA_HTTP_DIR}"/high-confidence_sINDEL_sSNV_in_HC_regions_v1.2.1.merged.vcf.gz
wget -P ${INPUT_DIR} "${DATA_HTTP_DIR}"/high-confidence_sINDEL_sSNV_in_HC_regions_v1.2.1.merged.vcf.gz.tbi

Running DeepSomatic with one command

DeepVariant pipeline consists of 3 steps: make_examples_somatic, call_variants, and postprocess_variants. You can run DeepSomatic with one command using the run_deepvariant script.

Running on a CPU-only machine

BIN_VERSION="1.8.0"

sudo docker pull google/deepsomatic:"${BIN_VERSION}"

sudo docker run \
-v ${INPUT_DIR}:${INPUT_DIR} \
-v ${OUTPUT_DIR}:${OUTPUT_DIR} \
google/deepsomatic:"${BIN_VERSION}" \
run_deepsomatic \
--model_type=WGS_TUMOR_ONLY \
--ref=${INPUT_DIR}/GCA_000001405.15_GRCh38_no_alt_analysis_set.chr1.fna \
--reads_tumor=${INPUT_DIR}/HCC1395_wgs.tumor.chr1.bam \
--output_vcf=${OUTPUT_DIR}/HCC1395_deepsomatic_output.vcf.gz \
--sample_name_tumor="HCC1395Tumor" \
--num_shards=$(nproc) \
--logging_dir=${OUTPUT_DIR}/logs \
--intermediate_results_dir=${OUTPUT_DIR}/intermediate_results_dir \
--use_default_pon_filtering=true \
--regions=chr1

By using --use_default_pon_filtering=true the somatic variants will be filtered using the default PON vcf that contains variant calls from dbSNP, gnomAD and 1000 genomes. If you plan to customize post-filtering, then you can set this parameter to false and use custom filtering.

NOTE: If you want to run each of the steps separately, add --dry_run=true to the command above to figure out what flags you need in each step. Based on the different model types, different flags are needed in the make_examples step.

--intermediate_results_dir flag is optional. By specifying it, the intermediate outputs of make_examples_somatic and call_variants stages can be found in the directory.

sudo docker pull pkrusche/hap.py:latest
# Run hap.py
sudo docker run \
-v ${INPUT_DIR}:${INPUT_DIR} -v ${OUTPUT_DIR}:${OUTPUT_DIR} \
pkrusche/hap.py:latest \
/opt/hap.py/bin/som.py \
-N ${INPUT_DIR}/high-confidence_sINDEL_sSNV_in_HC_regions_v1.2.1.merged.vcf.gz \
${OUTPUT_DIR}/HCC1395_deepsomatic_output.vcf.gz \
-r ${INPUT_DIR}/GCA_000001405.15_GRCh38_no_alt_analysis_set.chr1.fna \
-o ${OUTPUT_DIR}/sompy_output/deepsomatic.chr1.sompy.output \
--feature-table generic \
-R ${INPUT_DIR}/High-Confidence_Regions_v1.2.bed \
-l chr1

The output:

      type  total.truth  total.query    tp    fp   fn  unk  ambi    recall  recall_lower  recall_upper   recall2  precision  precision_lower  precision_upper  na  ambiguous  fp.region.size   fp.rate
0   indels          133          368    86   282   47    0     0  0.646617      0.562921      0.724009  0.646617   0.233696         0.192656         0.278905   0          0       248956422  1.132728
1     SNVs         3440         4813  2937  1876  503    0     0  0.853779      0.841676      0.865287  0.853779   0.610222         0.596381         0.623931   0          0       248956422  7.535455
5  records         3573         5181  3023  2158  550    0     0  0.846068      0.833956      0.857619  0.846068   0.583478         0.570011         0.596852   0          0       248956422  8.668184