Merge pull request #11 from MPUSP/dev

feat: various improvements to variant effect prediction + viz

Author: m-jahn

.test/config/config.yml

@@ -57,13 +57,21 @@ variant_calling:
     extra: ""
   freebayes:
     extra: "--ploidy 1 --min-base-quality 20 --min-alternate-fraction 0.75"
-  effect_prediction:
+
+variant_annotation:
+  tool: "vep"
+  vep:
     convert_gff: True
     plugins: []
     extra: "--everything --species custom_bacteria --distance 0"
+  snpeff:
+    extra: "-nodownload -ud 0"
 
 qc:
   multiqc:
     extra: "--verbose --dirs"
   fastqc:
     extra: "--quiet"
+
+report:
+  minimum_variant_count: 2

README.md

@@ -23,7 +23,7 @@ Detailed information about input data and workflow configuration can also be fou
 
 If you use this workflow in a paper, don't forget to give credits to the authors by citing the URL of this repository or its DOI.
 
-*Workflow overview:*
+_Workflow overview:_
 
 <!-- include overview-->
 <img src="resources/images/dag.png" align="center" />

config/README.md

@@ -1,6 +1,10 @@
 ## Workflow overview
 
 This workflow is a best-practice workflow for mapping of reads to reference genomes, minimalistic and simple.
+
+It will attempt to map reads to the reference using one of the included mappers, report read and experiment statistics, create coverage profiles, quantify variants (such as SNPs) using two different tools, and predict the effect of these variants.
+All of this is performed with minimal input and without lookups to external databases (e.g. for variant effects), which makes the workflow ideal for bacteria and other low-complexity non-model organisms.
+
 The workflow is built using [snakemake](https://snakemake.readthedocs.io/en/stable/) and consists of the following steps:
 
 1. Download genome reference from NCBI (`ncbi tools`), or use manual input (`fasta`, `gff` format)
@@ -13,8 +17,9 @@ The workflow is built using [snakemake](https://snakemake.readthedocs.io/en/stab
 5. Determine experiment type, get mapping stats (`rseqc`)
 6. Generate `bigwig` or `bedgaph` coverage profiles (`deeptools`)
 7. Quantify variations and SNPs (`bcftools`, `freebayes`)
-8. Predict effect of variants such as premature stop codons (`VEP`)
-9. Collect statistics from tool output (`MultiQC`)
+8. Predict effect of variants such as premature stop codons (`VEP` or `SnpEff`)
+9. Create consensus of variants and create a visual report (`R markdown`)
+10. Collect statistics from tool output (`MultiQC`)
 
 ## Running the workflow
 
@@ -82,10 +87,14 @@ This table lists all parameters that can be used to run the workflow.
 | extra                    | string  | additional arguments to BCFtools filter                  |                   |
 | _freebayes_              |         |                                                          |                   |
 | extra                    | string  | additional arguments to Freebayes call                   |                   |
-| _effect_prediction_      |         |                                                          |                   |
+| **variant_annotation**   |         |                                                          |                   |
+| tool                     | string  | annotation tool to use, one of 'vep', 'snpeff'           | `vep`             |
+| _vep_                    |         |                                                          |                   |
 | convert_gff              | boolean | whether to convert NCBI GFF to Ensemble style GFF        | `True`            |
 | plugins                  | array   | VEP plugins to use                                       | `[]`              |
 | extra                    | string  | additional arguments to VEP                              | see config.yml    |
+| _snpeff_                 |         |                                                          |                   |
+| extra                    | string  | additional arguments to SnpEff                           | see config.yml    |
 | **qc**                   |         |                                                          |                   |
 | _fastqc_                 |         |                                                          |                   |
 | extra                    | string  | additional arguments to FastQC                           |                   |

config/config.yml

@@ -57,13 +57,21 @@ variant_calling:
     extra: ""
   freebayes:
     extra: "--ploidy 1 --min-base-quality 20 --min-alternate-fraction 0.75"
-  effect_prediction:
+
+variant_annotation:
+  tool: "vep"
+  vep:
     convert_gff: True
     plugins: []
     extra: "--everything --species custom_bacteria --distance 0"
+  snpeff:
+    extra: "-nodownload -ud 0"
 
 qc:
   multiqc:
     extra: "--verbose --dirs"
   fastqc:
     extra: "--quiet"
+
+report:
+  minimum_variant_count: 2

config/schemas/config.schema.yml

@@ -1,4 +1,4 @@
-$schema: "http://json-schema.org/draft-07/schema#"
+$schema: "https://json-schema.org/draft/2020-12/schema"
 description: configuration file for the analysis pipeline
 properties:
   samplesheet:
@@ -153,7 +153,19 @@ properties:
           extra:
             type: string
             description: additional arguments to pass to Freebayes
-      effect_prediction:
+    required:
+      - bcftools_pileup
+      - bcftools_call
+      - bcftools_view
+      - bcftools_filter
+      - freebayes
+  variant_annotation:
+    type: object
+    properties:
+      tool:
+        type: string
+        description: annotation tool to use, one of 'vep', 'snpeff'
+      vep:
         type: object
         properties:
           convert_gff:
@@ -167,13 +179,17 @@ properties:
           extra:
             type: string
             description: additional arguments to pass to VEP
+      snpeff:
+        type: object
+        properties:
+          extra:
+            type: string
+            description: additional arguments to pass to SnpEff
     required:
-      - bcftools_pileup
-      - bcftools_call
-      - bcftools_view
-      - bcftools_filter
-      - freebayes
-      - effect_prediction
+      - tool
+      - vep
+      - snpeff
+
   qc:
     type: object
     properties:
@@ -192,11 +208,19 @@ properties:
     required:
       - multiqc
       - fastqc
+  report:
+    type: object
+    properties:
+      minimum_variant_count:
+        type: integer
+        description: minimum number of samples/appearances for the same variant to be included in the final report
 required:
   - samplesheet
   - get_genome
   - fastp
   - mapping
   - mapping_stats
   - variant_calling
+  - variant_annotation
   - qc
+  - report

config/schemas/samples.schema.yml

@@ -1,4 +1,4 @@
-$schema: "http://json-schema.org/draft-07/schema#"
+$schema: "https://json-schema.org/draft/2020-12/schema"
 description: entries for the sample sheet
 properties:
   sample:

workflow/Snakefile

@@ -14,6 +14,7 @@ include: "rules/bcftools.smk"
 include: "rules/freebayes.smk"
 include: "rules/variant_annotation.smk"
 include: "rules/qc.smk"
+include: "rules/report.smk"
 
 
 # optional messages, log and error handling
@@ -35,4 +36,5 @@ onerror:
 rule all:
     input:
         "results/multiqc/multiqc_report.html",
+        expand("results/report/{caller}_report.pdf", caller=["bcftools", "freebayes"]),
     default_target: True

workflow/envs/bcftools.yml

@@ -0,0 +1,7 @@
+name: bcftools
+channels:
+  - conda-forge
+  - bioconda
+  - nodefaults
+dependencies:
+  - bcftools=1.22

workflow/envs/report_html.yml

@@ -0,0 +1,16 @@
+name: report_html
+channels:
+  - conda-forge
+  - bioconda
+  - nodefaults
+dependencies:
+  - r-base 4.3.1
+  - r-essentials=4.3
+  - r-tidyverse=2.0.0
+  - r-ggrepel=0.9.3
+  - r-rstatix=0.7.2
+  - r-ggpubr=0.6.0
+  - r-ggupset=0.4.0
+  - bioconductor-biostrings=2.68.1
+  - bioconductor-genomicranges=1.52.0
+  - bioconductor-genomicfeatures=1.52.1

workflow/envs/report_pdf.yml

@@ -0,0 +1,8 @@
+name: report_pdf
+channels:
+  - conda-forge
+  - bioconda
+  - nodefaults
+dependencies:
+  - python=3.9
+  - weasyprint=62.3