Which type of feature are you implementing?
Is there an existing issue / PR for this?
Is there already an implementation for this feature?
Describe the feature
This feature adds a new Nextflow DSL2 pipeline, nf-UPDhmm, for the detection of uniparental disomy (UPD) events in trio sequencing data. The workflow integrates the R/Bioconductor package UPDhmm with preprocessing modules specifically designed for clinical genomic datasets.
Background
UPDs are rarely assessed routinely in genetic diagnostics, despite their relevance in explaining certain diseases, as existing methods have significant limitations in detection and characterization. Currently, there is no standardized, containerized workflow for automated UPD analysis in trio sequencing data. This feature aims to implement a reproducible Nextflow DSL2 pipeline following nf-core guidelines, ensuring compatibility across HPC and cloud environments.****
Proposed implementation
Default steps:
Preprocessing (PREPROCESS_VCF)
Prepares the trio data by combining individual VCFs (if required) and applying filters to keep only high-quality biallelic SNVs.
-
REMOVE_ANNOTATIONS - Remove annotations from individual VCFs and regroup by family.
-
COMBINE_VCF - Perform VCF intersection and merge operations. If params.perform_intersection is true, only shared variants among trio members are retained. Otherwise, all variants from the trio are combined directly.
-
SV_MASK_BED - Filter structural variants using BED regions.
-
FILTER_LOW_CONF - Apply hard filters to produce the final preprocessed VCF per family including:
1) Keep only biallelic sites
2) Exclude variants where all trio members are homozygous reference (0/0 or 0|0)
3) Enforce minimum genotype quality (GQ) and depth (DP) across the trio
4) Exclude problematic genomic regions (centromeres, segmental duplications, HLA/KIR)
UPD detection (UPDHMM ANALISYS)
Runs the UPDhmm core functions to detect genomic blocks consistent with UPD.
- vcfCheck – Validates and formats the combined VCF as a largeCollapsedVcf object.
- calculateEvents – Applies the HMM (Viterbi algorithm) to infer hidden states, groups variants into blocks, and annotates each block with confidence metrics.
- collapseEvents - Merges overlapping blocks of the same UPD type within a chromosome.
Anything else?
No response
Related Issues
No response
Are you going to work on this?
Which type of feature are you implementing?
Labelfor the featureIs there an existing issue / PR for this?
Is there already an implementation for this feature?
Describe the feature
This feature adds a new Nextflow DSL2 pipeline, nf-UPDhmm, for the detection of uniparental disomy (UPD) events in trio sequencing data. The workflow integrates the R/Bioconductor package UPDhmm with preprocessing modules specifically designed for clinical genomic datasets.
Background
UPDs are rarely assessed routinely in genetic diagnostics, despite their relevance in explaining certain diseases, as existing methods have significant limitations in detection and characterization. Currently, there is no standardized, containerized workflow for automated UPD analysis in trio sequencing data. This feature aims to implement a reproducible Nextflow DSL2 pipeline following nf-core guidelines, ensuring compatibility across HPC and cloud environments.****
Proposed implementation
Default steps:
Preprocessing (PREPROCESS_VCF)
Prepares the trio data by combining individual VCFs (if required) and applying filters to keep only high-quality biallelic SNVs.
REMOVE_ANNOTATIONS - Remove annotations from individual VCFs and regroup by family.
COMBINE_VCF - Perform VCF intersection and merge operations. If
params.perform_intersectionis true, only shared variants among trio members are retained. Otherwise, all variants from the trio are combined directly.SV_MASK_BED - Filter structural variants using BED regions.
FILTER_LOW_CONF - Apply hard filters to produce the final preprocessed VCF per family including:
1) Keep only biallelic sites
2) Exclude variants where all trio members are homozygous reference (0/0 or 0|0)
3) Enforce minimum genotype quality (GQ) and depth (DP) across the trio
4) Exclude problematic genomic regions (centromeres, segmental duplications, HLA/KIR)
UPD detection (UPDHMM ANALISYS)
Runs the UPDhmm core functions to detect genomic blocks consistent with UPD.
Anything else?
No response
Related Issues
No response
Are you going to work on this?
Assigneesto facilitate tracking who is working on this.