Getting Started

# download and build
git clone https://github.com/huangnengCSU/longcallR.git
cd longcallR
cargo build --release # The executable will be located at: target/release/longcallR

# call SNPs and phasing
longcallR -b input.bam -f ref.fa -o output -t 8 -p ont-cdna       # Nanopore cDNA reads
longcallR -b input.bam -f ref.fa -o output -t 8 -p ont-drna       # Nanopore dRNA reads (no strand bias filtering)
longcallR -b input.bam -f ref.fa -o output -t 8 -p hifi-isoseq    # PacBio iso-seq reads
longcallR -b input.bam -f ref.fa -o output -t 8 -p hifi-masseq    # PacBio mas-seq reads (no strand bias filtering)

# Note: If read strandedness is uncertain, set --strand-bias false to avoid over-filtering.
longcallR -b input.bam -f ref.fa -o output -t 8 -p <preset> --strand-bias false

# Allele-specific junction analysis
longcallR asj -a annotation.gtf.gz -b phased.bam -f ref.fa -o output_prefix -t threads

# Allele-specific expression analysis
longcallR ase -a annotation.gtf.gz -b phased.bam -o output_prefix -t threads

Table of Contents

• Introduction
• Installation
• Usage
• Demo
• Alignment
• Citation
• Update Log
• License

Introduction

longcallR is a tool for SNP calling, haplotype phasing, and allele-specific analysis with long-read RNA-seq data. It supports ONT cDNA and direct RNA sequencing, as well as PacBio Iso-Seq and MAS-Seq.

Full documentation is available at huangnengCSU.github.io/longcallR.

Installation

Install from source

LongcallR is written in Rust and uses Cargo to build as follows:

git clone https://github.com/huangnengCSU/longcallR.git
cd longcallR
cargo build --release

INSTALL_DIR="$(pwd)/target/release"
export PATH="$PATH:$INSTALL_DIR"

# Check Installation
longcallR -h

Install from Conda

longcallR can be installed via Conda:

mamba install longcallr

# Check Installation
longcallR -h

Install from Crates.io

longcallR can be installed from crates.io:

cargo install longcallR

# Check Installation
longcallR -h

Usage

Basic Example

longcallR \
--bam-path input.bam \                  ## The alignment bam file
--ref-path ref.fa \                     ## The reference file must be indexed.
--preset ${PRESET} \                    ## option: {ont-cdna, ont-drna, hifi-isoseq, hifi-masseq}
--output ${OUTPUT_PREFIX}               ## Prefix of output files

For advanced options, see the documentation

Advanced option: if a phased VCF is already available, --direct-haplotag can be used with --input-vcf to directly haplotag reads without re-phasing SNPs.

Allele specific analysis

# call allele-specific expression
longcallR ase -b <phased_bam> -a <annotation> -o <output_prefix> -t <threads>

# call allele-specific junction
longcallR asj -b <phased_bam> -a <annotation> -f <reference> -o <output_prefix> -t <threads>

# store allele-specific junction in BED format for IGV visualization
asj_to_bed.py output.asj.tsv [p_value_threshold] > output.asj.bed

For detailed information on available options and output file formats, please refer to the documentation

Demo

longcallR -b demo/demo.bam -f demo/chr20.fa -o demo/test -t 8 -p hifi-masseq

Alignment

Based on the strand orientation of reads from different PacBio and Nanopore protocols, we recommend the following alignment parameters for minimap2:

minimap2 -ax splice:hq -uf ref.fa query.fa > aln.sam    # PacBio Kinnex/Mas-seq
minimap2 -ax splice:hq ref.fa query.fa > aln.sam        # PacBio Iso-seq
minimap2 -ax splice -uf -k14 ref.fa reads.fa > aln.sam  # Nanopore dRNA
minimap2 -ax splice ref.fa reads.fa > aln.sam           # Nanopore cDNA

Note: The -uf option forces minimap2 to consider only a single transcript strand during alignment.

• PacBio Kinnex/MAS-Seq and Nanopore dRNA sequencing produce single-stranded reads.
• Nanopore cDNA sequencing produces double-stranded reads.
• PacBio Iso-Seq datasets may be either single- or double-stranded.

We recommend using the -uf option for single-stranded reads and omitting it for double-stranded reads.

Results:

The results of manuscript can be found and reproduced using the data and scripts available at: Zenodo.

Citation

If you use LongcallR in your work or analysis, please cite the preprint:

Neng Huang, Heng Li, Human Pangenome Reference Consortium, SNP calling, haplotype phasing and allele-specific analysis with long RNA-seq reads. nature methods, 2026. https://doi.org/10.1038/s41592-026-03045-6

Update Log

2.0.1 - 2026-04-03

• Improved robustness of --direct-haplotag mode: non-SNV records (indels, spanning deletions, symbolic alleles) in the input VCF are now filtered out before haplotagging to prevent mismatched interpretation at indel loci.

2.0.0 - 2026-03-27

• Added support for direct haplotagging from a pre-phased VCF using --input-vcf --direct-haplotag.
• Reimplemented the functionality of longcallR-ase.py and longcallR-asj.py in Rust and integrated them into longcallR as the longcallR ase and longcallR asj subcommands. The Python implementations will be removed in a future release.

License

MIT License

Copyright (c) 2024 Dana-Farber Cancer Institute.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.