isoquant 3.13.1

Reference-based analysis and quantification of long RNA reads

Full IsoQuant documentation can be found here. Information in this README is given only for convenience and is not a full user manual.

Current version: see VERSION file.

• Citation information
• Feedback and bug reports
• Quick start examples

About IsoQuant

IsoQuant is a tool for the genome-based analysis of long RNA reads, such as PacBio or Oxford Nanopores. IsoQuant allows reconstructing and quantifying transcript models with high precision and decent recall. If the reference annotation is given, IsoQuant also assigns reads to the annotated isoforms based on their intron and exon structure. IsoQuant further performs annotated gene, isoform, exon, and intron quantification. If reads are grouped (e.g. according to a cell type), counts are reported according to the provided grouping.

The latest IsoQuant version can be downloaded from github.com/ablab/IsoQuant/releases/latest.

Full IsoQuant documentation is available at ablab.github.io/IsoQuant.

Supported sequencing data

IsoQuant supports all kinds of long RNA data:

• PacBio CCS
• ONT dRNA / ONT cDNA
• Assembled / corrected transcript sequences

Reads must be provided in FASTQ/FASTA format (can be gzipped) or unmapped BAM format. If you have already aligned your reads to the reference genome, simply provide sorted and indexed BAM files. IsoQuant expect reads to contain polyA tails. For more reliable transcript model construction do not trim polyA tails.

IsoQuant can also take aligned Illumina reads to correct long-read spliced alignments. However, short reads are not used to discover transcript models or compute abundances.

Supported reference data

Reference genome is mandatory and should be provided in multi-FASTA format (can be gzipped).

Reference gene annotation is not mandatory but is likely to increase precision and recall. It can be provided in GFF/GTF format (can be gzipped).

Pre-constructed minimap2 index can also be provided to reduce mapping time.

Citation

The paper describing IsoQuant algorithms and benchmarking is available at 10.1038/s41587-022-01565-y.

To try IsoQuant, you can use the data that was used in the publication zenodo.org/record/7611877.

Feedback and bug reports

Your comments, bug reports, and suggestions are very welcome. They will help us to further improve IsoQuant. If you have any troubles running IsoQuant, please send us isoquant.log from the <output_dir> directory.

You can leave your comments and bug reports at our GitHub repository tracker or send them via email: [e-mail].

Quick start

• Full IsoQuant documentation is available at ablab.github.io/IsoQuant.

• IsoQuant can installed via pip:

  pip install isoquant
  

• Via conda (bioconda channel):

  conda create -c conda-forge -c bioconda -n isoquant python=3.12 isoquant
  

• Or from GitHub:

  git clone https://github.com/ablab/IsoQuant.git 
  cd IsoQuant
  git checkout latest
  pip install -e .
  

Installation typically takes no more than a few minutes.

• If running simply from the source archive, you will need Python3 (3.8 or higher), gffutils, pysam, biopython, pyfaidx, ssw-py, editdistance and some other common Python libraries to be installed. See requirements.txt for details. You will also need to have minimap2 and samtools to be in your $PATH variable. All required Python libraries can be installed via:

  pip install -r requirements.txt
  

• Verify your installation by running (typically takes less than 1 minute):

  isoquant --test
  

• To run IsoQuant on raw FASTQ/FASTA files, use the following command

  isoquant --reference /PATH/TO/reference_genome.fasta \
  --genedb /PATH/TO/gene_annotation.gtf \
  --fastq /PATH/TO/sample1.fastq.gz /PATH/TO/sample2.fastq.gz \
  --data_type (assembly|pacbio_ccs|nanopore) -o OUTPUT_FOLDER
  

  For example, using the toy data provided within this repository,

  isoquant --fastq /home/andreyp/ablab/IsoQuant/isoquant_tests/simple_data/chr9.4M.ont.sim.fq.gz \
  --reference /home/andreyp/ablab/IsoQuant/isoquant_tests/simple_data/chr9.4M.fa.gz \
  --genedb /home/andreyp/ablab/IsoQuant/isoquant_tests/simple_data/chr9.4M.gtf.gz \
  --data_type nanopore --complete_genedb -p TEST_DATA --output isoquant_test 
  

• To run IsoQuant on aligned reads (make sure your BAM is sorted and indexed) use the following command:

    isoquant --reference /PATH/TO/reference_genome.fasta \
    --genedb /PATH/TO/gene_annotation.gtf \
    --bam /PATH/TO/sample1.sorted.bam /PATH/TO/sample2.sorted.bam \
    --data_type (assembly|pacbio_ccs|nanopore) -o OUTPUT_FOLDER
  

• If using official annotations containing gene and transcript features use --complete_genedb to save time.

• Using reference annotation is optional since version 3.0, you may preform de novo transcript discovery without providing --genedb option':

    isoquant --reference /PATH/TO/reference_genome.fasta \
    --fastq /PATH/TO/sample1.fastq.gz /PATH/TO/sample2.fastq.gz \
    --data_type (assembly|pacbio|nanopore) -o OUTPUT_FOLDER
  

• If multiple files are provided, IsoQuant will create a single output annotation and a single set of gene/transcript expression tables.