Tools & Scientific References
Peer-reviewed publications for the bioinformatics tools, databases, and guidelines that power the ATGC Flow pipeline
See how these methods are applied in our platform & pipeline specs, or start from the fundamentals in the Genomics Knowledge Base.
Standards and Guidelines for the Interpretation of Sequence Variants (ACMG/AMP 2015)
The American College of Medical Genetics and Genomics and the Association for Molecular Pathology jointly developed updated standards and guidelines for the clinical interpretation of sequence variants. This document establishes the 5-tier classification system (Pathogenic, Likely Pathogenic, VUS, Likely Benign, Benign) with evidence criteria (PVS1, PS1–PS4, PM1–PM6, PP1–PP5, BA1, BS1–BS4, BP1–BP7) that underpins all modern variant interpretation, including the automated ACMG classification engine in ATGC Flow.
All References
Standards and Guidelines for the Interpretation of Sequence Variants (ACMG/AMP 2015)
The American College of Medical Genetics and Genomics and the Association for Molecular Pathology jointly developed updated standards and guidelines for the clinical interpretation of sequence variants. This document establishes the 5-tier classification system (Pathogenic, Likely Pathogenic, VUS, Likely Benign, Benign) with evidence criteria (PVS1, PS1–PS4, PM1–PM6, PP1–PP5, BA1, BS1–BS4, BP1–BP7) that underpins all modern variant interpretation, including the automated ACMG classification engine in ATGC Flow.
A framework for variation discovery and genotyping using next-generation DNA sequencing data (GATK)
Describes the Genome Analysis Toolkit (GATK) framework for discovery and genotyping of SNPs and indels from next-generation sequencing data. Introduces the concept of Base Quality Score Recalibration (BQSR) and the HaplotypeCaller variant caller. GATK 4.x is the variant calling backbone of the ATGC Flow pipeline, implementing HaplotypeCaller in GVCF mode for germline variant discovery.
Fast and accurate short read alignment with Burrows-Wheeler Aligner (BWA-MEM)
Presents BWA-MEM, the short-read alignment algorithm that maps sequences against a large reference genome with low divergence. BWA-MEM supports gapped alignment for short-read paired-end sequencing and achieves better performance than alternative algorithms. BWA-MEM2 (the successor) is used in ATGC Flow for aligning FASTQ reads to the GRCh38 reference genome.
ANNOVAR: Functional annotation of genetic variants from high-throughput sequencing data
Introduces ANNOVAR, an efficient software tool for functional annotation of genetic variants detected from diverse genomes including human exome and genome sequencing. Supports gene-based, region-based and filter-based annotation strategies using locally downloaded databases. ATGC Flow uses ANNOVAR to annotate variants with gene function, population frequencies (gnomAD), pathogenicity scores (dbNSFP), and clinical significance (ClinVar).
The Ensembl Variant Effect Predictor (VEP)
Describes VEP, Ensembl's variant annotation and effect prediction tool. VEP determines the effect of variants (SNPs, insertions, deletions, CNVs, structural variants) on genes, transcripts, and protein sequence, as well as regulatory regions. VEP v113 is integrated into the ATGC Flow pipeline for HGVS nomenclature generation, consequence prediction, and population frequency annotation.
The mutational constraint spectrum quantified from variation in 141,456 humans (gnomAD)
Presents the Genome Aggregation Database (gnomAD) v2.1 containing 125,748 exomes and 15,708 genomes from unrelated individuals. Provides allele frequency data across 8 population groups used to filter common variants in disease studies. ATGC Flow uses gnomAD v4.1 (807,162 exomes + 76,215 genomes) as the primary allele frequency filter for rare variant prioritization, applying the PM2 ACMG criterion for variants absent from gnomAD.
fastp: an ultra-fast all-in-one FASTQ preprocessor
Presents fastp, a tool for quality control, adapter trimming, and filtering of FASTQ files. fastp is 2–4× faster than Trimmomatic, automatically detects adapter sequences for paired-end data, and generates built-in JSON and HTML QC reports. ATGC Flow uses fastp as the first pipeline step for quality control and adapter trimming before alignment.
Nextflow enables reproducible computational workflows
Describes Nextflow, a workflow system for creating scalable, portable, and reproducible computational pipelines. Nextflow DSL2 supports modular pipeline design and seamless execution across local, HPC, and cloud environments using Docker/Singularity containers. ATGC Flow's entire bioinformatics pipeline is built with Nextflow DSL2, enabling reproducible execution from FASTQ to annotated VCF.
A universal SNP and small-indel variant caller using deep neural networks (DeepVariant)
Presents DeepVariant, a variant caller that uses a convolutional neural network trained on pileup images of aligned reads to classify SNPs and indels. DeepVariant achieves >99.9% F1 score on SNPs and outperforms GATK HaplotypeCaller on indels in several benchmarks. ATGC Flow supports DeepVariant as an alternative variant caller for Illumina WES data.
dbNSFP v4: a comprehensive database of transcript-specific functional predictions and annotations for human nonsynonymous and splice-site SNVs
Describes dbNSFP v4, a database that compiles functional predictions and annotations from 33 algorithms (SIFT, PolyPhen-2, CADD, REVEL, MetaSVM, etc.) and allele frequencies from multiple population databases for all potential nonsynonymous and splice-site SNVs in the human genome. ATGC Flow incorporates dbNSFP v4.7 to provide in silico pathogenicity evidence for the PP3 and BP4 ACMG criteria.
ClinVar: improving access to variant interpretations and supporting evidence
Describes ClinVar, NCBI's freely accessible archive of human variants and their relationships to human health. ClinVar aggregates submissions from clinical laboratories, research groups, and expert panels and provides classifications with supporting evidence. ATGC Flow queries ClinVar to apply PP5 (reputable source reports variant as pathogenic) and BP6 (reputable source reports variant as benign) ACMG evidence criteria.
Open-Source Tools Powering ATGC Flow
Every component of the pipeline is built on peer-reviewed, community-validated bioinformatics software
Acknowledge ATGC Flow
If ATGC Flow contributed to your research or analysis, please acknowledge the software as: