Variant analysis is now a key tool in biomedical research. Even so, we know that working with complex genomic data is not always simple. At Biotechvana we want to accompany you in that process, acting as a trusted partner between massive sequencing and biological interpretation, with genotyping services that are solid, transparent, and designed for the people behind each project.
Every study raises different questions, and bioinformatics should not be a one-size-fits-all solution. That is why we adapt our workflows to your specific needs: from whole-exome and whole-genome analyses for Mendelian studies, to pipelines tailored for cancer research. Likewise, our experience covers amplicon processing and capture systems, ensuring maximum precision for both global analyses and targeted panels.
Our main value is personalization. We analyze your experimental design in detail and carefully adjust detection and annotation parameters to deliver reliable, clear results ready to be interpreted in their clinical or basic-research context.
This workflow is designed to identify germline variants associated with hereditary diseases or specific genetic traits. Our approach combines exhaustive data quality control with rigorous analysis of familial inheritance models, whether through trio studies or complex pedigrees. Through deep functional annotation, we prioritize variants with true pathogenic impact, transforming genomic data into biologically relevant results that precisely identify the underlying genetic cause.
- Quality analysis of sequencing reads (FASTQ) using standard metrics.
- Read preprocessing, including demultiplexing and removal of low-quality sequences, primer/adapter remnants, and artifacts.
- Read curation when appropriate.
- Alignment of processed reads against the reference genome/transcriptome.
- Generation of alignment metrics: mapping percentage, mean coverage, coverage uniformity, and off-target reads (in exomes).
- Marking and removal of PCR duplicates to avoid biases in variant quantification.
- Base quality score recalibration (BQSR) to adjust empirical quality scores and reduce systematic false positives.
- Alignment refinement around insertion/deletion regions (indels).
- Calling SNP and indel variants using high-sensitivity germline algorithms (e.g., HaplotypeCaller).
- Joint genotyping in cohorts or family trios to increase statistical power.
- Detection of structural variants (CNVs, large deletions/duplications) tailored to exome or genome data.
- Comprehensive annotation with population databases (gnomAD, 1000G) and clinical databases (ClinVar, OMIM, dbSNP).
- Filtering by technical quality (read depth, genotype quality, strand bias).
- Filtering by population frequency: selection of rare variants (<1% or <5% depending on phenotype).
- Biological prioritization: family segregation analysis (dominant, recessive, de novo, X-linked models).
- Pathogenicity classification according to ACMG guidelines.
This protocol addresses the biological complexity of tumors through a bioinformatics approach designed to manage heterogeneity and sample purity. The analysis focuses on precise differentiation between somatic and germline variants, enabling identification of driver mutations even at low allele frequencies. In addition to SNP and indel detection, we evaluate tumor mutational burden (TMB), mutational signatures, and structural or copy-number alterations (CNVs), providing an integrated view of the tumor genomic landscape.
- Creation of a Panel of Normals (PON) to identify and filter recurring sequencing artifacts.
- Comprehensive quality control of FASTQ files (Tumor and Normal when available).
- Rigorous read cleaning to reduce false positives, especially for low-frequency mutations.
- Alignment against the reference genome.
- Inference of mapping performance metrics.
- Duplicate marking.
- Base recalibration (BQSR), a critical step to detect variants with low allele frequency (VAF).
- Somatic variant calling (SNVs and indels) by comparing Tumor-Normal pairs, or in tumor-only mode when required by the experimental design.
- Application of a Panel of Normals (PON) to filter technical noise and recurring sequencing artifacts.
- Estimation of tumor purity and ploidy.
- Analysis of somatic copy-number variations (CNVs) and loss of heterozygosity (LOH).
- Detection of complex structural variants (translocations, inversions).
- Exclusion of germline variants to isolate tumor-exclusive mutations.
- Filtering by variant allele frequency (VAF) and coverage depth.
- Annotation with reference oncology databases (COSMIC, CIViC, OncoKB).
- Identification of driver versus passenger variants.
- Reporting of biomarkers with predictive or prognostic value (e.g., tumor mutational burden - TMB, microsatellite instability - MSI).
This protocol is aimed at achieving maximum depth and precision in restricted genomic regions, making it ideal for targeted panel studies or variant validation. Thanks to its high sensitivity, it is especially effective at detecting low-frequency variants, which are essential for mosaicism analysis or liquid biopsy projects. The workflow ensures uniform coverage and superior statistical reliability, ensuring that every read contributes to a robust and reproducible result in the regions of greatest interest.
- Read quality analysis.
- Strict removal of primer and adapter sequences specific to the capture/amplicon kit to avoid false positives at read ends.
- Read merging in the case of overlapping amplicons when required by the design.
- Restricted alignment focused on target regions (ROI - Regions of Interest).
- On-target performance metrics: percentage of reads aligned within the panel design versus flanking regions.
-
Duplicate handling:
- In hybrid capture: marking and removal of duplicates (PCR dedup).
- In amplicons: management of biological vs technical duplicates (or use of UMIs - Unique Molecular Identifiers when present for error correction).
- Local realignment around indels.
- Variant calling with parameters tuned for high depth.
- Detection of low-frequency variants (mosaicism or subclonality) thanks to high coverage.
- Functional annotation of variants found within panel regions.
- Strict filtering by minimum coverage (e.g., >100x or >500x depending on clinical/research context).
- Strand-bias analysis to discard PCR errors.
- Validation of specific variants requested by the client.
- Report focused on the genes of interest from the designed panel.