Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/nanoporetech/bonito
A PyTorch Basecaller for Oxford Nanopore Reads
https://github.com/nanoporetech/bonito
basecalling nanopore pytorch
Last synced: about 2 months ago
JSON representation
A PyTorch Basecaller for Oxford Nanopore Reads
- Host: GitHub
- URL: https://github.com/nanoporetech/bonito
- Owner: nanoporetech
- License: other
- Created: 2019-11-19T13:46:22.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2023-12-12T17:17:26.000Z (7 months ago)
- Last Synced: 2024-05-09T15:34:30.291Z (about 2 months ago)
- Topics: basecalling, nanopore, pytorch
- Language: Python
- Homepage: https://nanoporetech.com/
- Size: 745 KB
- Stars: 374
- Watchers: 36
- Forks: 116
- Open Issues: 93
-
Metadata Files:
- Readme: README.md
Lists
- awesome-nanopore - Bonito - [Python] - A PyTorch Basecaller for Oxford Nanopore Reads (research, not production basecaller) (Software packages / Basecalling)
README
# Bonito
[](https://badge.fury.io/py/ont-bonito)
[](https://img.shields.io/badge/python-3.8-brightgreen.svg)
[](https://img.shields.io/badge/python-3.9-brightgreen.svg)
[](https://img.shields.io/badge/python-3.10-brightgreen.svg)
[](https://img.shields.io/badge/python-3.11-brightgreen.svg)
[](https://img.shields.io/badge/cuda-11.7-blue.svg)
Bonito is an open source research basecaller for Oxford Nanopore reads.
For anything other than basecaller training or method development please use [dorado](https://github.com/nanoporetech/dorado).
```bash
$ pip install --upgrade pip
$ pip install ont-bonito
$ bonito basecaller [email protected] /data/reads > basecalls.bam
```
Bonito supports writing aligned/unaligned `{fastq, sam, bam, cram}`.
```bash
$ bonito basecaller [email protected] --reference reference.mmi /data/reads > basecalls.bam
```
Bonito will download and cache the basecalling model automatically on first use but all models can be downloaded with -
``` bash
$ bonito download --models --show # show all available models
$ bonito download --models # download all available models
```
## Modified Bases
Modified base calling is handled by [Remora](https://github.com/nanoporetech/remora).
```bash
$ bonito basecaller [email protected] /data/reads --modified-bases 5mC --reference ref.mmi > basecalls_with_mods.bam
```
See available modified base models with the ``remora model list_pretrained`` command.
## Training your own model
To train a model using your own reads, first basecall the reads with the additional `--save-ctc` flag and use the output directory as the input directory for training.
```bash
$ bonito basecaller dna_r9.4.1 --save-ctc --reference reference.mmi /data/reads > /data/training/ctc-data/basecalls.sam
$ bonito train --directory /data/training/ctc-data /data/training/model-dir
```
In addition to training a new model from scratch you can also easily fine tune one of the pretrained models.
```bash
bonito train --epochs 1 --lr 5e-4 --pretrained [email protected] --directory /data/training/ctc-data /data/training/fine-tuned-model
```
If you are interested in method development and don't have you own set of reads then a pre-prepared set is provide.
```bash
$ bonito download --training
$ bonito train /data/training/model-dir
```
All training calls use Automatic Mixed Precision to speed up training. To disable this, set the `--no-amp` flag to True.
## Developer Quickstart
```bash
$ git clone https://github.com/nanoporetech/bonito.git # or fork first and clone that
$ cd bonito
$ python3 -m venv venv3
$ source venv3/bin/activate
(venv3) $ pip install --upgrade pip
(venv3) $ pip install -e .
```
## Interface
- `bonito view` - view a model architecture for a given `.toml` file and the number of parameters in the network.
- `bonito train` - train a bonito model.
- `bonito evaluate` - evaluate a model performance.
- `bonito download` - download pretrained models and training datasets.
- `bonito basecaller` - basecaller *(`.fast5` -> `.bam`)*.
### References
- [Sequence Modeling With CTC](https://distill.pub/2017/ctc/)
- [Quartznet: Deep Automatic Speech Recognition With 1D Time-Channel Separable Convolutions](https://arxiv.org/pdf/1910.10261.pdf)
- [Pair consensus decoding improves accuracy of neural network basecallers for nanopore sequencing](https://www.biorxiv.org/content/10.1101/2020.02.25.956771v1.full.pdf)
### Licence and Copyright
(c) 2019 Oxford Nanopore Technologies Ltd.
Bonito is distributed under the terms of the Oxford Nanopore
Technologies, Ltd. Public License, v. 1.0. If a copy of the License
was not distributed with this file, You can obtain one at
http://nanoporetech.com
### Research Release
Research releases are provided as technology demonstrators to provide early access to features or stimulate Community development of tools. Support for this software will be minimal and is only provided directly by the developers. Feature requests, improvements, and discussions are welcome and can be implemented by forking and pull requests. However much as we would like to rectify every issue and piece of feedback users may have, the developers may have limited resource for support of this software. Research releases may be unstable and subject to rapid iteration by Oxford Nanopore Technologies.