prRONTo - precise rRNA modification analysis on ONTechnoligies

prRONTo is a snakemake workflow for comprehensive identification of diverse ribosomal RNA modifications by targeted nanopore direct RNA sequencing and JACUSA2.

See original publication for details DOI: 10.1080/15476286.2023.2248752. Please cite this publication if you use prRONTo in your work.

Installation

Chekout the repository and install required packages (see conda-lock.yaml).

git clone https://github.com/dieterich-lab/prRONTo
cd prRONTo

We recommend to use Conda or for faster dependecy resolving Mamba to install the required packages.

Requirements

Internally, prRONTo uses JACUSA2 and JACUSA2helper to detect RNA modifications and a collection of R and Python scripts for processing.

Required dependencies can be installed via Conda:

conda env create -n pronto -f conda-lock.yaml
conda activate pronto

For faster dependecy resolving, we highly recommend to use Mamba. Refer to Mamba Installation to setup an mamba instance and install required dependencies with:

mamba env create -n pronto -f conda-lock.yaml
mamba activate pronto

Usage

A a sample description via a PEP file and a sample table is required. Furthermore, the reference FASTA and a customized RNA modification file corresponding to the sequencing data is required.

snakemake -c 1 --snakefile <SNAKEFILE> --config pepfile=<PEPFILE> [--configfile=<CONFIG_FILE>]

The <CONFIG_FILE> defines and adjusts parameters of the analysis whereas the <PEPFILE> is entirely sample specific.

PEP file

The PEP is in yaml format. In the following, a descriptive example is presented:

pep_version: 2.0.0
sample_table: <SAMPLE_TABLE>      # REQUIRED, path to sample table

project: "Example"                # OPTIONAL, plain text, no special characters

pronto:
  regions: ["region1", "region2"] # REQUIRED, list of seqnames to scan for modifications
  output: <OUTPUT_DIRECTOR>       # REQUIRED, path where results will be written to
  mods: <MODS_FILE>               # REQUIRED, path to existing modification annotation
  ref: <REF_FASTA>                # REQUIRED, path to reference FASTA

  # values for condition 1 and 2 
  # must exist in
  # the column "condition" in the sample table!
  condition1: "condition1"        # REQUIRED, value from col. "condition" in sample table
  condition2: "condition2"        # REQUIRED, value from col. "condition", in sample table

See example/human/pep.yaml for an example.

Sample table

A minimal sample table is provided in the following:

sample_name	condition	bam
sample_1_1	condition1	<BAM_1_1>
sample_1_2	condition1	<BAM_1_2>
sample_2_1	condition2	<BAM_2_1>
sample_2_2	condition2	<BAM_2_2>

See example/human/sample_table.yaml for an example.

Analysis config

In the following, a descriptive example is presented:

downsampling:   # OPTIONAL
  # target coverage
  reads: [1000, ]
  # seeds for read sampling with samtools
  # the number of seeds corresponds to the number of repetitions
  seed: ["CfdCY", "gJm9e", "CZ8X7", "1Cdq4", "6pod1", "RtDnQ", "AdOWe", ]
jacusa2:
  features: ["M", "MDI"]    # Features to use for the LOF analysis
# LOF specific parameters
lof:
  - neighbors: 20
    contamination: 0.001
  - neighbors: 20
    contamination: 0.002

Modification file format

A tab-delimeted file with three columns: seq_id, pos(0-indexed), and mod.

In the following a section from examples/data/human_rrna.tsv:

seq_id	pos	mod
NR_003286_RNA18SN5	26	Am
NR_003286_RNA18SN5	33	psU
NR_003286_RNA18SN5	35	psU
NR_003286_RNA18SN5	92	psU
...	...	...

Report

The report is organised in to 4 sections:

• Results
• Read
• Modifications and
• Session.

Results

The results section contains the putative RNA modifications as outliers stratified by region and utilized feature. Additionaly, known modifications, optional downsampling info and LOF parameters are presented.

Reads

This section is mainly based on samtools stats|coverage. Properties of reads and summary statistics are presented:

• total reads,
• mapped reads,
• read length,
• mapping quaylity,
• ...

Modifications

Summary of provided RNA modification file.

Session

Paths to config files and installed software can be checked in this section.

Example

In the following, a chain of commands to run a toy example:

1. Clone repository: git clone https://github.com/dieterich-lab/prRONTo
2. cd prRONTo
3. Install conda env.: conda env create -n pronto -f conda-lock.yaml
4. Activate conda env.: conda activate pronto
5. Run prRONTo: `cd example/human ; run_exampe.sh``
6. Open output/report/report.htmlwith your favorite browser and check the results.