Germline Build WGS gCNV Model
Implementation of the helper_gcnv_model_wgs
step
The helper_gcnv_model_wgs
step takes as the input the results of the ngs_mapping
step
(aligned germline reads) and builds a model that can be used by GATK4 gCNV. Important: the workflow
assumes that all samples in the cohort use the same library kit and all are WGS.
Step Input
The step uses Snakemake sub workflows for the result of the ngs_mapping
(aligned reads BAM files).
Step Output
All donors will be used to generate the two parts of the required gCNV model, specifically:
ploidy-model
and cnv_calls-model
. Both are required to execute gCNV in CASE mode.
For example, the relevant directories might look as follows:
work/
+-- bwa.gcnv_contig_ploidy.default
`-- out
`-- bwa.gcnv_contig_ploidy.default
|-- SAMPLE_0
| |-- contig_ploidy.tsv
| |-- global_read_depth.tsv
| |-- mu_psi_s_log__.tsv
| |-- sample_name.txt
| `-- std_psi_s_log__.tsv
|-- [...]
`-- bwa.gcnv_contig_ploidy.default
`-- ploidy-model
|-- contig_ploidy_prior.tsv
|-- gcnvkernel_version.json
|-- interval_list.tsv
|-- mu_mean_bias_j_lowerbound__.tsv
|-- mu_psi_j_log__.tsv
|-- ploidy_config.json
|-- std_mean_bias_j_lowerbound__.tsv
`-- std_psi_j_log__.tsv
+-- bwa.gcnv_call_cnvs.default.***_of_***
`-- out
`-- bwa.gcnv_call_cnvs.default.***_of_***
|-- cnv_calls-calls
| |-- SAMPLE_0
| `-- [...]
| |-- [...]
|-- cnv_calls-model
| |-- denoising_config.json
| |-- gcnvkernel_version.json
| |-- interval_list.tsv
| |-- log_q_tau_tk.tsv
| |-- mu_W_tu.tsv
| |-- mu_ard_u_log__.tsv
| |-- mu_log_mean_bias_t.tsv
| |-- mu_psi_t_log__.tsv
| |-- std_W_tu.tsv
| |-- std_ard_u_log__.tsv
| |-- std_log_mean_bias_t.tsv
| `-- std_psi_t_log__.tsv
`-- cnv_calls-tracking
`-- [...]
Global Configuration
At the moment, no global configuration is used.
Default Configuration
The default configuration is as follows.
step_config:
helper_gcnv_model_wgs:
#path_ngs_mapping: ../ngs_mapping
gcnv: # REQUIRED
# Path to interval block list with PAR region for contig calling.
# path_par_intervals: ''
#
# path to BED file with uniquely mappable regions.
path_uniquely_mapable_bed: # REQUIRED