ENH: add docs for output.

[brentp]

Is your feature request related to a problem? Please describe.

After running, there is a lot of output! Where to begin or how to orient would be useful.

Describe the solution you'd like

A section in the readme on navigating the output would be helpful.

Additional context

Hi, the pipeline ran very smoothly! Thanks for that.
When finished, I see:

$ ls results/_LAST/out
bam_statistics           joint_small_variants_vcf        mitorsaw_vcf_index                pharmcat_report_html            small_variant_gvcf                       tertiary_small_variant_compound_het_vcf_index
bcftools_roh_bed         joint_small_variants_vcf_index  mosdepth_depth_distribution_plot  pharmcat_report_json            small_variant_gvcf_index                 tertiary_small_variant_filtered_tsv
bcftools_roh_out         joint_sv_vcf                    mosdepth_region_bed               phase_blocks                    small_variant_stats                      tertiary_small_variant_filtered_vcf
cpg_combined_bed         joint_sv_vcf_index              mosdepth_region_bed_index         phase_haplotags                 snv_distribution_plot                    tertiary_small_variant_filtered_vcf_index
cpg_combined_bed_index   joint_trgt_vcf                  mosdepth_summary                  phase_stats                     stats_file                               tertiary_sv_filtered_tsv
cpg_combined_bw          joint_trgt_vcf_index            msg_file                          phased_small_variant_vcf        sv_copynum_bedgraph                      tertiary_sv_filtered_vcf
cpg_hap1_bed             mapq_distribution_plot          paraphase_output_json             phased_small_variant_vcf_index  sv_copynum_summary                       tertiary_sv_filtered_vcf_index
cpg_hap1_bed_index       merged_haplotagged_bam          paraphase_realigned_bam           phased_sv_vcf                   sv_depth_bw                              trgt_coverage_dropouts
cpg_hap1_bw              merged_haplotagged_bam_index    paraphase_realigned_bam_index     phased_sv_vcf_index             sv_gc_bias_corrected_depth_bw            trgt_spanning_reads
cpg_hap2_bed             methbat_profile                 paraphase_vcfs                    phased_trgt_vcf                 sv_maf_bw                                trgt_spanning_reads_index
cpg_hap2_bed_index       mg_distribution_plot            pbstarphase_json                  phased_trgt_vcf_index           sv_supporting_reads
cpg_hap2_bw              mitorsaw_hap_stats              pharmcat_match_json               read_length_plot                tertiary_small_variant_compound_het_tsv
indel_distribution_plot  mitorsaw_vcf                    pharmcat_phenotype_json           read_quality_plot               tertiary_small_variant_compound_het_vcf

It would be nice to have more instruction on where to focus

[williamrowell]

Completely agree, and we're working on improving documentation. Does this help at all, for a place to start?

[brentp]

Ah, yes. I missed that. Thanks very much, Billy. Additional docs would be appreciated, but that gives me a good starting point.

Another feature request would be that the first job in the workflow goes through all files in tertiary_map and ref_map and checks for existence. I have a few long paths that I messed up and the error didn't manifest until late in the pipeline. It was able to continue once I fixed, but nice to have any simple errors like this out of the way.

[williamrowell]

Building input validation tasks in WDL isn't straightforward, as all task inputs have to be localized to the job node, which isn't a big deal on HPC but could get time consuming in the cloud (imagine localizing all input BAMs for a trio or quad to a single disk in AWS just to validate that the paths are correct).

I'm planning a change that should further simplify providing static inputs, and hopefully that addresses part of the problem. I'll keep thinking about ways to quickly validate the sample inputs.