SIAV-VariantFlow is a reproducible bioinformatics workflow for the detection, annotation and visualization of low-frequency intrahost genetic variation in swine influenza A virus (swIAV) from Illumina paired-end sequencing data.
The workflow covers the analysis from raw sequencing reads to annotated variant tables and coverage visualizations.
Example dataset: The sequencing data used in this repository correspond specifically to the H1N2 study associated with NCBI BioProject PRJNA994299.
SIAV-VariantFlow grew out of the bioinformatics workflows I progressively developed during my doctoral research at IRTA-CReSA and the Universitat Autònoma de Barcelona (UAB) to investigate the within-host evolution of swine influenza A virus.
These analyses evolved across studies involving H1N1, H3N2 and H1N2 viruses under different immune scenarios, including vaccination, previous infection and coinfection. SIAV-VariantFlow consolidates and updates these analyses into a documented and reproducible workflow that can be adapted to other influenza A virus sequencing datasets.
NCBI SRA / Illumina paired-end reads
│
▼
00_download_data.sh
│
▼
Raw FASTQ files
│
▼
01_quality_control.sh
│
FastQC / Trimmomatic
/ MultiQC
│
▼
Trimmed paired reads
│
├──────────────────────────┐
▼ │
02_build_inoculum_consensus.sh │
│ │
▼ │
Inoculum consensus genome │
│ │
▼ │
04_analyze_samples.sh ◄──────────────┘
│
Mapping / BAM processing
Coverage calculation
BQSR / LoFreq
Variant filtering
SnpEff annotation
│
┌─────┴─────┐
▼ ▼
all_variants.tsv all_depth.tsv
│ │
▼ ▼
05_prepare_variant 06_plot_coverage.R
_table.sh
│ │
▼ ▼
Annotated variants Coverage figures
The custom SnpEff database required for variant annotation is generated with:
03_build_snpeff_database.sh
Create the main Conda environment:
conda env create -f environment.yml
conda activate siav-variantflowSnpEff is maintained in a separate environment because it requires a newer Java runtime:
conda env create -f environment_snpeff.ymlThe complete H1N2 example can then be reproduced sequentially:
conda activate siav-variantflow
bash scripts/00_download_data.sh
bash scripts/01_quality_control.sh
bash scripts/02_build_inoculum_consensus.sh
conda activate siav-snpeff
bash scripts/03_build_snpeff_database.sh
conda activate siav-variantflow
bash scripts/04_analyze_samples.sh --all
bash scripts/05_prepare_variant_table.sh
Rscript scripts/06_plot_coverage.RSample identifiers and SRA accessions are provided in metadata/samples.tsv, while experimental groups used for visualization are defined in metadata/experimental_groups.tsv.
The R visualization step requires ggplot2, dplyr and ggpubr.
The core workflow maps sequencing reads against the reconstructed H1N2 inoculum consensus, calculates genome-wide sequencing depth, performs base-quality recalibration, calls low-frequency variants with LoFreq, applies the filtering criteria used in the associated study and annotates retained variants using a custom SnpEff database.
Default SNV criteria:
Minimum sequencing depth: 100 reads
Minimum alternative read count: 10 reads
LoFreq additionally performs statistical variant-quality filtering.
These thresholds reflect the criteria implemented for the associated analysis and should be independently evaluated when applying the workflow to other sequencing protocols or biological questions.
The workflow generates three main analysis-ready outputs:
results/summary/all_depth.tsv
results/summary/all_variants.tsv
results/summary/all_variants_annotated.tsv
Although the example included here reproduces the H1N2 study, the workflow originated from analyses developed across several swine influenza A virus experiments.
For other influenza A virus datasets, users should adapt the sample metadata, reference genome and annotation, SnpEff database, experimental groups and, where scientifically justified, variant-filtering thresholds.
Particular attention should be paid to reference and annotation consistency because of the segmented influenza A genome and overlapping gene products.
SIAV-VariantFlow was developed and refined from bioinformatics analyses used across the following swine influenza A virus studies:
López-Valiñas Á, et al. (2021).
Identification and Characterization of Swine Influenza Virus H1N1 Variants Generated in Vaccinated and Nonvaccinated, Challenged Pigs.
Viruses 13:2087.
DOI: 10.3390/v13102087
López-Valiñas Á, et al. (2022).
Evolution of Swine Influenza Virus H3N2 in Vaccinated and Nonvaccinated Pigs after Previous Natural H1N1 Infection.
Viruses 14:2008.
DOI: 10.3390/v14092008
López-Valiñas Á, et al. (2023).
Vaccination against swine influenza in pigs causes different drift evolutionary patterns upon swine influenza virus experimental infection and reduces the likelihood of genomic reassortments.
Frontiers in Cellular and Infection Microbiology 13:1111143.
DOI: 10.3389/fcimb.2023.1111143
López-Valiñas Á, et al. (2023).
Genetic diversification patterns in swine influenza A virus (H1N2) in vaccinated and nonvaccinated animals.
Frontiers in Cellular and Infection Microbiology 13:1258321.
DOI: 10.3389/fcimb.2023.1258321
The sequencing dataset provided as the reproducible example in this repository corresponds to the H1N2 study above (BioProject PRJNA994299).
If you use SIAV-VariantFlow, please cite this repository and the publication most relevant to the analysis being reproduced.
For analyses based on the H1N2 example dataset, please also cite:
López-Valiñas Á, Valle M, Pérez M, Darji A, Chiapponi C, Ganges L, Segalés J, Núñez JI. Genetic diversification patterns in swine influenza A virus (H1N2) in vaccinated and nonvaccinated animals. Frontiers in Cellular and Infection Microbiology. 2023;13:1258321. doi:10.3389/fcimb.2023.1258321.
A machine-readable citation for SIAV-VariantFlow is provided in CITATION.cff.
Álvaro López-Valiñas, PhD
Bioinformatics · Viral genomics · Infectious diseases
ORCID: 0000-0002-7492-5108
SIAV-VariantFlow originates from the bioinformatics analyses developed during my doctoral research on swine influenza A virus evolution.
López-Valiñas, Álvaro. Evolution of swine influenza virus associated with vaccination.
The complete PhD thesis is available through the Universitat Autònoma de Barcelona institutional repository:
https://ddd.uab.cat/pub/tesis/2023/hdl_10803_688403/alv1de1.pdf