QuasiFlow
Introduction
QuasiFlow is a nextflow pipeline for reproducible analysis of NGS-based HIVDR testing data across different computing environments. The pipeline takes raw sequence reads in FASTQ format as input, performs quality control, mapping of reads to a reference genome, variant calling, querying the database for detection of HIV drug resistance mutations, and ultimately generates a user-friendly report in PDF and HTML format. QuasiFlow is publicly available at https://github.com/AlfredUg/QuasiFlow.
Installation
QuasiFlow requires nextflow (version 21.04.3 or higher) and any of conda/singularity/docker. In this walk through, we shall demonstrate the use of conda which is more readily available to most users.
The first option is to install the pipeline using nextflow, it will be installed in the $HOME directory under the .nextflow sub-directory. Confirm that installation was successful by printing out the help message.
nextflow pull AlfredUg/QuasiFlow
nextflow run ~/.nextflow/assets/AlfredUg/QuasiFlow --helpAlternatively, simply clone the pipeline repository into a desired directory. Similarly, confirm that installation was successful by printing out the help message.
git clone https://github.com/AlfredUg/QuasiFlow.git
nextflow run QuasiFlow --helpUsage
The pipeline takes as input paired-end illumina data in FASTQ format. Let's download some test data from the European Nucleotide Archive (ENA) using wget command and decompress it using the gunzip command. This is paired-end data from a single sample of bioProject PRJDB3502.
wget ftp://ftp.sra.ebi.ac.uk/vol1/fastq/DRR030/DRR030218/DRR030218_1.fastq.gz
wget ftp://ftp.sra.ebi.ac.uk/vol1/fastq/DRR030/DRR030218/DRR030218_2.fastq.gz gunzip DRR030218*.gzRun QuasiFlow on a test dataset with default parameters under the conda profile. This option does not require prior installation since it automatically pulls the pipeline from main branch of the pipeline repository on github. In addition, it installs all the dependancies in a conda environment. If you already installed the pipeline using the procedure above, see next options.
nextflow run AlfredUg/QuasiFlow -r main --reads "$PWD/*_{1,2}.fastq" -profile condaIf you pulled/installed the pipeline using nextflow, simply point to the installation path as follows;
nextflow run ~/.nextflow/assets/AlfredUg/QuasiFlow --reads "$PWD/*_{1,2}.fastq" -profile condaSimilarly, if you already cloned the pipeline repository, simply point to the installation path as follows;
nextflow run path/to/QuasiFlow --reads "$PWD/*_{1,2}.fastq" -profile condaProfiles
Quasiflow can be run under different computing environments, simply choose an appropriate profile via the -profile argument. Could take any of the following -profile conda, singularity, docker. Custom profiles can be added to the conf directory using any of the available profiles as a template.
Pipeline
Outputs Quality control
raw_reads_multiqc_report.html: Aggregated quality control data and visualisations - one file for entire dataset
Variants and drug resistance outputs
consensus*.fasta:FASTAfiles of consensus sequences - one per sampleconsensus*.json:JSONfiles of detailed HIV drug resistance analysis - one per sampledr_report*.csv:CSVfiles of drug resistance mutations at different mutational frequencies - one per samplefiltered*.fastq:FASTQfiles of drug resistance mutations at different mutational frequencies - one per samplemutation_report*.aavf:AAVFfiles of amino acid variant calls - one per samplehivdr*.html:HTMLFinal drug resistance report - one per sample
Pipeline information output
QuasiFlow_DAG.html: Graphical representation of the pipeline's processes/operators and channels between them.QuasiFlow_report.html: Overall start and completion time, CPU and memory usage.QuasiFlow_timeline.html: Timeline for all the processes executed in the pipeline.
Note: Nextflow throws the following warning on MacOS, WARN: Task runtime metrics are not reported when using macOS without a container engine.
Parameters
HyDRA parameters
Mandatory parameters
--reads: Path to input data (must be surrounded with quotes)
Optional parameters
-
--reporting_threshold: Minimum mutation frequency percent to report. -
--consensus_pct: Minimum percentage a base needs to be incorporated into the consensus sequence. -
--min_read_qual: Minimum quality for a position in a read to be masked. -
length_cutoff: Reads which fall short of the specified length will be filtered out. -
score_cutoff: Reads that have a median or mean quality score (depending on the score type specified) less than the score cutoff value will be filtered out. -
--min_variant_qual: Minimum quality for variant to be considered later on in the pipeline. -
--min_dp: Minimum required read depth for variant to be considered later on in the pipeline. -
--min_ac: The minimum required allele count for variant to be considered later on in the pipeline -
--min_freq: The minimum required frequency for mutation to be considered in drug resistance report.
Sierralocal parameters
Optional parameters
-
--xml: Path to HIVdb ASI2 XML. -
--apobec-tsv: Path to tab-delimited (tsv) HIVdb APOBEC DRM file. -
--comments-tsv: Path to tab-delimited (tsv) HIVdb comments file.
Output parameters
Optional parameters
--outdir: Path to directory where results will be saved
Dependancies.
Below is the list of tools that are used in the QuasiFlow pipeline. These tools are readliy available and may be installed using conda via bioconda channel.
- fastQC
- MultiQC
- Trim-galore
- Quasitools
- Sierra-local
- R packages (Jsonlite, plyr, dplyr, flexdashboard, rmarkdown, knitr, tinytex), all available in CRAN
Troubleshooting
Kindly report any issues at https://github.com/AlfredUg/QuasiFlow/issues.
License
QuasiFlow is licensed under GNU GPL v3.
Citation
Ssekagiri A, Jjingo D, Lujumba I, Bbosa N, Bugembe DL, Kateete DP, Jordan IK, Kaleebu P, Ssemwanga D. QuasiFlow: a Nextflow pipeline for analysis of NGS-based HIV-1 drug resistance data. Bioinform Adv. 2022 Nov 28;2(1):vbac089. doi: 10.1093/bioadv/vbac089. PMID: 36699347; PMCID: PMC9722223.