Variant Calling Service

Overview

The Variant Calling Service detects genomic variations in individual genomes using high-throughput sequencing data. This provides molecular-level insights for disease research, biomedical research, and evolutionary biology studies. The methodology relies on a multi-step computational pipeline. First, sequence alignment tools map raw sequencing reads to a reference genome to generate BAM/SAM files. Subsequently, sorting, duplicate marking, and base quality recalibration processes eliminate technical artifacts. Finally, variant detection algorithms analyze processed alignment data to identify variations and generate standardized VCF files. Annotation tools then assess the biological impact of these variants. This service supports research workflows with high efficiency, accuracy, and standardization. It is bridging the gap between genotypes and phenotypes.

Our Variant Calling Service empowers your research with:

• High-Precision Variant Detection: Significantly reduce false positives, enhance rare variant identification, and ensure consistent genotyping across all samples.
• Advanced Biomarker Profiling: Detects candidate biomarkers.
• Robust Data Handling: Advanced algorithms efficiently process low-quality and low-depth sequencing data while minimizing batch effects and technical inconsistencies.
• Comprehensive Analysis Workflow: An integrated end-to-end framework encompassing quality assessment, reference genome alignment, variant calling, and functional annotation.​

What Is Variant Calling

Variant calling uses high-throughput sequencing. It fully sequences and studies DNA differences in individual or species-wide genomes. This process finds many genetic variations, including single nucleotide variants (SNVs), insertion/deletion sites (indels), structural variations (SVs) and copy number variations (CNVs). It checks every base of the whole genome. This creates essential data for later research. The results enable efficient comparative genomic studies, precise gene mapping and creation of molecular markers.

How to Measure

1. Sample Requirements

• Total DNA Amount: FFPE samples ≥250 ng, cfDNA samples ≥30 ng, ensuring sufficient quantity for library construction.
• DNA Integrity: FFPE fragments ≥1000 bp; cfDNA fragments 140–180 bp, preventing excessive fragmentation from affecting variant detection.
• DNA Purity: A260/280 (1.8–2.0) ; A260/230 (1.8–2.2) , excluding protein/organic contaminant interference.

2. Sequencing

Table 1: Comparative genotyping techniques

3. Quality Control

• Sample-level QC:
• Remove duplicate samples
• Exclude samples with low DNA concentration (<10 ng/μL)
• Identify and discard contaminated samples using control checks
• Data-level QC: Filter SNPs with: Missingness >20%; Hardy-Weinberg equilibrium p-value < 1×10⁻⁶ Minor allele frequency (MAF) <1%.
• Implement batch effect correction when applicable:
• Linear Regression: Fits a linear model to remove batch-related variation while preserving biological signals.
• ComBat: Empirical Bayesian adjustment that retains biological differences by estimating batch-specific parameters.
• MNN Correct: Matches mutual nearest neighbor cells across batches to compute batch correction vectors in low-dimensional space.

4. Reference Alignment or Assembly

High-quality sequencing reads are mapped to a standardized reference genome. Thus it enables the identification of genetic variations through comparison with the reference and accurate detection and interpretation of variants.

5. Statistical Analysis Workflow

Variant Calling

• Germline Variant Calling: Identifies inherited genetic variations encompassing SNVs and indels.

Application: Characterize hereditary disorders, population genetic structures, and familial disease susceptibility.

• Germline Joint Variant Calling: Performed across sample cohorts to enhance variant detection accuracy by leveraging shared genetic information.

Application: Improves sensitivity for low-frequency variants and mitigates false positives through unified statistical modeling. It is optimal for large-scale genomic studies and population-level analyses.

• Somatic Variant Calling: Detects acquired SNVs and indels in tumor samples, pinpointing mutations driving oncogenesis.

Application: Address challenges from tumor purity and subclonal heterogeneity, serving as a cornerstone for cancer genomics and cancer genomics research.

• Somatic Variant Calling with Germline Filtration: Distinguish somatic mutations from inherited germline variants by integrating matched normal samples or population databases.

Application: Reduce false positives by filtering germline polymorphisms and technical artifacts, enhancing specificity for tumor-specific alterations.

Copy Number Variations (CNVs) Analysis

CNV analysis identifies large-scale genomic alterations (>1 kb), including duplications or deletions of DNA segments, through statistical modeling of sequencing depth deviations and segmentation algorithms.

These variations drive pathogenic mechanisms in genetic disorders, developmental abnormalities, and neurodegenerative diseases.

Structural Variations (SV) Detection

SV detection characterizes large-scale genomic rearrangement, such as inversions, translocations, and insertions/deletions (>50 bp), by integrating discordant read pairs and split-read alignments from high-throughput sequencing data.

This analysis is critical for elucidating complex genetic disorders and oncogenic events.

Microsatellite Instability (MSI) Status

MSI refers to genetic alterations caused by defects in the DNA mismatch repair system. This leads to length variations in microsatellites (short repetitive DNA sequences, 1–6 nucleotides) through insertions or deletions during replication.

MSI is classified into MSI-High (MSI-H, instability in ≥2/5 markers ), MSI-Low (MSI-L, instability in one marker) and Microsatellite Stable (MSS, No instability). NGS-Based assays analyze hundreds of microsatellite loci and is applicable to FFPE tissue or ctDNA.

Research significance:

• Lynch syndrome–related research
• Associations with disease progression and chemotherapy response
• Associations with immunotherapy response
• Cancer Associations

Tumor Mutational Burden (TMB)

TMB measures the total number of genetic mutations found in the DNA of cancer cells. This is often reported as the number of mutations per megabase (Mb) of interrogated genomic sequence. In common practice, TMB thresholds are stratified as High TMB (TMB-H, ≥10 mut/Mb for pan-cancer applications) and Low TMB (TMB-L, <10 mut/Mb, indicating lower tumor immunogenicity and reduced likelihood of benefiting from ICIs).

Research applications:

• Immunotherapy Prediction
• Relationship with MSI
• Associations with treatment outcomes
• Prognostic Value

6. Annotation and Visualization

• Variant location and impact
• Variant frequencies in population studies
• IGV (Integrative Genomics Viewer): Dynamically visualizes BAM alignment files (e.g., sequencing depth, coverage), VCF variant loci (SNP/InDel/SV), CNV, and gene annotations.
• SCOUT: Automates the generation and visual presentation of genetic variant interpretation reports, integrating variant filtering and ACMG classification.
• PUZZLE: Provides mutation spectrum statistics (e.g., SNP/InDel distribution) and enables rapid screening of high-frequency disease-associated mutations.
• MyVCF: Generates mutation heatmaps and functional enrichment analysis (e.g., GO/KEGG pathways) for multi-sample genomic comparisons.
• Circos: Uses circular layouts to display chromosomal structural variations (e.g., translocations, inversions), CNV, and gene fusion events.

Figure 1: Variant Calling Service Workflow

What Can We do

Data Acquisition & Preprocessing

• Extensive Data Sourcing: Access genomic, morphological, and ecological data from public repositories (e.g., GenBank, Dryad) and private collaborations, spanning diverse species from microbes to macroorganisms.
• Stringent Quality Control: Implement rigorous data filtering to remove errors, correct sequences, and guarantee high-quality inputs.

Germline vs. Somatic Variant Analysis

• Germline Variants: Focus on heritable variants associated with genetic disorders or cancer predisposition syndromes.
• Somatic Variants: Filter germline variants using population databases; Prioritize tumor driver mutations for mechanistic studies and outcome associations.

Tumor-Specific Analysis

• Tumor Mutational Burden (TMB) Calculation: Requires filtering deamination artifacts in FFPE samples to avoid false positives.
• Automated Pipelines: Platforms like Cancer Genomics Cloud (CGC) support somatic variant calling.

Value-Added Services

• Project Integration: Hereditary disease research (with family-based validation) and interpretation of mutations relevant to therapeutic research.
• Deep Learning Optimization: Enhanced sensitivity via hybrid models.

Customized Analysis & Reporting

• Tailored Workflows: Design analytical pipelines to address specific questions.
• Interactive Visualization & Reporting: Generate clear, publication-ready figures and detailed reports explaining methods, results, and implications, with recommendations for follow-up studies.
• Ongoing Support & Collaboration: Provide post-analysis support, additional analyses, and collaborative opportunities to help you refine interpretations and expand your research impact.

Our Advantages.

• Pipeline Customization: Tailor analytical workflows to your specific research objectives, whether reconstructing deep evolutionary histories, investigating recent speciation events, or testing hypotheses about adaptive evolution.
• Efficient Automation: Leverage automated, validated pipelines to accelerate analysis while minimizing human error, ensuring reproducibility and consistency across projects.
• Timely Delivery: Prioritize rapid turnaround without sacrificing analytical rigor, meeting urgent deadlines for grant proposals, conference presentations, or manuscript submissions.
• Expert Support: Benefit from bioinformaticians and evolutionary biologists who guide study design, troubleshoot challenges, and interpret results in the context of your research question.

Applications

Research into Human Disease Genetics and Risk

Variant calling enables the identification of disease-associated genetic mutations through high-throughput sequencing workflows. This process detects SNVs and INDELs in human genomes, supporting studies of hereditary disorders and pharmacogenomic analyses of drug-response variability.

Agricultural & Plant Breeding

In plant breeding, variant calling speeds up trait improvement. It finds genetic changes linked to valuable crop traits.

For hybrid crops, variant discovery in parent lines guides crosses. This adds beneficial gene versions while removing harmful ones.

Population, Quantitative, and Evolutionary Genetics

Variant calling underpins analyses of genetic diversity and evolutionary forces across populations. Population genetics leverages SNV/INDEL datasets to compute allele frequencies, detect selection signatures, and model demographic histories.

In quantitative genetics, variant calling enables genomic prediction of complex traits. It estimates heritability via mixed models and calculates kinship coefficients to correct for relatedness in GWAS.

Evolutionary studies integrate ancient DNA variant calling with methods like Treemix to reconstruct gene flow events and identify adaptive mutations.

Systematics and taxonomy

Variant calling resolves taxonomic classifications and phylogenetic relationships by comparing genomic variations across species. DNA barcoding relies on SNV profiling of conserved loci to delineate species boundaries. While whole-genome variant data (SNPs/INDELs) constructs cladograms to infer evolutionary divergence timelines. For cryptic species, variant density analysis identifies regions under divergent selection, clarifying classification ambiguities.

Ancestry and Kinship Analyses

Kinship and ancestry studies measure how closely people are related. They do this using variant data to calculate IBD probabilities and mixed ancestry proportions.

Tools like PLINK use SNV data to create kinship matrices. They also perform PCA to fix population bias in disease studies. Forensic applications leverage variant calling to verify familial relationships.

Demo

Figure 2: Number of overlapping variants after identifying multi-caller clusters for deletions, duplications, and mCNVs (Jakubosky, 2020)

Case Study

The complete sequence of a human genome

Journal：Science

Published：2022

• Background
• Methods
• Results

FAQs