Small Variant Interpretation
🔹 Multi-Variant Analysis¶
VarXOmics supports multi-variant analysis for small variants (SNPs & INDELs), copy number variants (CNVs), and structural variants (SVs).
This module allows users to upload VCF files, configure advanced filtering parameters, and perform variant interpretation in a structured and reproducible manner.
1. Input Data¶
- From the Home page, choose “Multi Variant”.
- Drop and upload a VCF file.
- Click Run to proceed to the variant filtering configuration window.
2. Variant Filtering Settings¶
After uploading the VCF, a configuration panel titled “Multi-Variant Analysis Settings” will appear.
This panel allows fine-tuning of variant filtration and prioritization parameters.
🧩 Sample Filters¶
| Setting | Description |
|---|---|
| Reference Genome | Select the genome build used for alignment — GRCh37 or GRCh38. |
| Gender | Choose the biological sex of the sample (Male, Female, or Unknown) for filtering X/Y-linked variants. |
| Only PASS Variants | When set to Yes, filters out non-PASS variants from the VCF. |
| Inheritance Pattern | Specify the expected mode of inheritance (Dominant, Recessive, Compound Heterozygous, or Any pattern). |
| HPO Terms (CSV) | Provide one or multiple HPO IDs (comma-separated) to integrate phenotype-based filtering, e.g. HP:0001250,HP:0000707. |
🌍 Allele Frequency Filters¶
| Setting | Description |
|---|---|
| AF ClinVar | Sets a frequency cap for ClinVar-derived evidence. Variants above this threshold are excluded. |
| AF (Predicted variants) | Defines the allele frequency cutoff for predicted variants (e.g., 0.05). Lower thresholds retain rarer variants. |
🤖 Predictive Scores¶
| Score | Description | Typical Cutoff |
|---|---|---|
| ADA | ADA score for splicing impact prediction. | 0.6 |
| RF | Random Forest model score for splicing variant prediction. | 0.6 |
| REVEL | REVEL score for missense variant pathogenicity prediction. | 0.75 |
🧬 SpliceAI Thresholds¶
| Parameter | Description | Default |
|---|---|---|
| Acceptor Loss (AL) | Minimum threshold for acceptor site loss prediction. | 0.5 |
| Acceptor Gain (AG) | Minimum threshold for acceptor site gain prediction. | 0.5 |
| Donor Loss (DL) | Minimum threshold for donor site loss prediction. | 0.5 |
| Donor Gain (DG) | Minimum threshold for donor site gain prediction. | 0.5 |
🧮 BayesDel Scores¶
| Parameter | Description | Default |
|---|---|---|
| BayesDel (with AF) | Bayesian deleteriousness score integrating allele frequency. | 0.069 |
| BayesDel (no AF) | Bayesian deleteriousness score excluding frequency adjustment. | -0.057 |
These thresholds align with backend default values for consistent filtering.
🧾 Classification Filters¶
AlphaMissense (AM) Classification¶
Select which ACMG/AMP pathogenicity tiers to retain in the analysis:
- Pathogenic
- Likely Pathogenic
- Ambiguous
- Benign
AlphaMissense (AM) Pathogenicity¶
- Numerical cutoff representing aggregated pathogenicity evidence (e.g., 0.564).
ClinVar Status¶
Select ClinVar review statuses to include:
- Pathogenic
- Likely Pathogenic
- Uncertain Significance
- Conflicting Classifications
- Benign
- Likely Benign
ACMG Classification¶
Specify ACMG evidence tiers for filtering:
- Pathogenic
- Likely Pathogenic
- Uncertain Significance
- Benign
- Likely Benign
3. Run Analysis¶
Once all filters are configured, click Start Analysis to execute the variant prioritization pipeline.
4. Result Overview¶
VarXOmics processes and prioritizes all variants based on its integrated filtering and scoring strategy. The final results are divided into six major sections, each providing an in-depth view of the analysis outcome.
🧬 Variant Overview¶
This section displays all filtered variants prioritized by the VarXOmics strategy. Each variant is annotated comprehensively, including genomic, transcript, clinical, and functional evidence.
Key Features:
- Detailed Annotation Table:
Displays complete variant-level information.
Users can customize visible columns using the “Columns” button.
- Quick Summary Subpanels:
Provides concise visual summaries of:
- Variant Consequence Distribution
- Variant Types
- Clinical Pathogenicity
- Affected Genes
These visual overviews allow users to rapidly assess the biological and clinical relevance of the prioritized variants.
📊 Variant Summary¶
This section provides an integrated summary of all filtered variants through seven visual plots, allowing users to explore the overall variant landscape.
Displayed Figures:
- SNP Density Plot:
Visualizes the distribution of SNPs across all chromosomes.
- ClinVar Ideogram:
Shows the chromosomal positions of ClinVar-annotated variants and their pathogenicity status.
- ClinVar Pathogenicity:
Summarizes the proportion of variants categorized by ClinVar pathogenicity.
- Variant Class:
Displays the proportion of SNPs, INDELs, and other variant types.
- Top 10 Disease Accumulators:
Highlights the diseases most frequently associated with the filtered variants.
- Top 10 Gene Accumulators:
Shows the genes where variants are most enriched.
- Variant Consequence Bar Plot:
Displays the frequency of different variant consequences (e.g., missense, nonsense, synonymous).
Together, these plots offer an intuitive view of the biological and clinical impact of the variant set.
🔬 Enrichment Analysis¶
This section performs functional enrichment for all genes linked to the filtered variants, helping users interpret the biological processes and pathways involved.
Components:
- GO Term Enrichment Table:
Lists significantly enriched Gene Ontology (GO) terms (Biological Process, Molecular Function, Cellular Component).
- KEGG Pathway Table:
Shows KEGG pathway enrichment results for the filtered gene set.
- GO Enrichment Visualization:
- Top enriched GO terms ranked by gene count.
- GO Term Counts.
- Statistics of GO Biological Processes, Molecular Functions, Cellular Components.
- KEGG Enrichment Visualization:
Presents top pathways ranked by enrichment significance.
The enrichment section provides direct insight into the biological themes underlying the filtered variant set.
📚 GWAS & QTL Associations¶
This section presents variant- and gene-level associations collected from multiple public resources, allowing users to explore known evidence linking the queried variants or genes to complex traits, diseases, and regulatory effects.
Contents cover: GWAS, eQTL, pQTL, Pharmacogenomics (PGx).
Users can customize visible columns with the “Columns” button (top-right corner).
Together, these resources allow users to assess the translational and regulatory significance of their filtered variants.
🕸️ Network¶
This section visualizes all biological interactions derived from the filtered variant and gene set.
This integrative network provides a global view of how variants, genes, proteins, drugs, and phenotypes interconnect through multiple evidence layers.
This network provides an integrated, evidence-weighted visualization of how variants influence genes, pathways, and drug responses — bridging molecular biology and clinical relevance.
Interactive Features:
- Edge Weight Filter:
Adjust the slider to dynamically filter edges weights. For each interactioins, the weights are set as:
| Interaction Type | Weight / Significance Cutoff |
|---|---|
| PPIs | Interaction confidence > 0.7 |
| GWAS | p < 5 × 10⁻⁸ |
| eQTL / pQTL | p_beta < 0.05 |
| MR (Mendelian Randomization) | p < 0.05 |
- Search Box:
Locate nodes by label (e.g., gene symbol, variant ID, drug name). - Node Summary Panel:
Displays total counts of nodes and edges, categorized by node and interaction type. - Table View:
Click the table icon in the bottom-right corner to open a structured view of the network, allowing users to inspect and filter interaction data directly.
✅ Tip:
Users can combine interaction filters and weight thresholds to create focused sub-networks, such as gene–drug pharmacogenomic networks or variant–phenotype associations for specific pathways.
🧠 Exomiser¶
VarXOmics integrates Exomiser analysis to support phenotype-driven variant prioritization. By linking variant annotations with HPO terms, Exomiser scores potential candidate variants and genes according to their phenotypic relevance.
This section presents:
- Ranked list of candidate genes and variants associated with input HPO phenotypes.
- Exomiser scores indicating phenotype similarity and variant pathogenicity.
- Cross-references with ClinVar and other annotations for interpretation consistency.
5. Export results¶
Users can export all results, figures, and enrichment tables as a single compressed archive by clicking “Export All” in the top-right corner of the interface.