Skip to content

Latest commit

 

History

History
175 lines (120 loc) · 6.13 KB

README.md

File metadata and controls

175 lines (120 loc) · 6.13 KB

Caution

DAJIN is deprecated. Please use DAJIN2, which has become the successor to DAJIN.

日本語はこちら

DAJIN is a genotyping software for genome-edited organisms using Nanopore sequencer.
DAJIN is named after 一網打尽 (Ichimou DAJIN or Yī Wǎng Dǎ jìn), meaning catching all in one net.

Features

  • Capturing mutations from SNV to SV, such as a point mutation, knock-out, knock-in, and inversion.
  • Automatic allele clustering and annotation.
  • Genotyping ~100 samples with different genome-editing designs at a single run.

Setup

We recommend Linux OS with NVIDIA GPU to reduce computation time.
If you use a Windows PC with NVIDIA GPU, please follow this instruction.

We confirmed DAJIN's operation on these environments.

1. Install conda

# Install miniconda
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
chmod +x Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh -b -f -p /usr/local/

2. Clone this repository

git clone /~https://github.com/akikuno/DAJIN.git

3. Prepare design files and fastq directory

We recommend the following directory tree. You can rename design.txt, design.fasta and fastq.

├── DAJIN
├── design.txt
├── design.fasta
├── fastq
│   ├── barcode01.fastq
│   ├── barcode02.fastq
│   ├── barcode03.fastq
│   ├── ......

Prepare inputs

design.txt

design.txt is formatted as below:

design=DAJIN/example/example.fa
input_dir=DAJIN/example/fastq
control=barcode01
grna=CCTGTCCAGAGTGGGAGATAGCC,CCACTGCTAGCTGTGGGTAACCC
genome=mm10
output_dir=DAJIN_example
threads=10
  • design: PATH to a FASTA file including sequences of each genotype. ">wt" and ">target" must be included.
  • input_dir: PATH to a directory containing demultiplexed FASTQ files
  • control: control barcode ID
  • grna: gRNA sequence(s) with PAM. multiple gRNA sequences must be delimitated by comma.
  • genome(optional): reference genome. e.g. hg38,mm10 (not GRCh38,GRCm38)
  • output_dir (optional): output directory name. Default is DAJIN_results
  • threads (optional: integer): Default is 2/3 of available CPU threads.
  • filter (optional: on or off): set filter to remove very minor allele (less than 3%). Default is on.

design, input_dir, control, genome,grna are required, but there are in no particular order.

design.fasta

design.fasta is a multi-FASTA file, which contains a WT and target sequence, as well as byproducts.

This is an example file of flox design.
In flox design, 6 allele types (WT, Target, Left LoxP, Right LoxP, flox deletion, and Inversion) may be produced.
DAJIN defines SV (structural variants) allele that is different from these alleles.

Usage

./DAJIN/DAJIN -i design.txt

Example usage

You can test DAJIN by an example small dataset.

./DAJIN/DAJIN -i DAJIN/example/design.txt

Outputs

DAJIN outputs two files and two folders: Details.csv, Details.pdf, Consensus, BAM.

Details.csv

Details.csv contains allele information. The allele with target mutation is labeled + in the Design column.

Sample Allele ID % of reads Allele type Indel Large indel Design
barcode01 1 100 wt - - -
barcode02 1 11.8 SV + + -
barcode02 2 88.2 target - - +
barcode03 1 9.9 SV + + -
barcode03 2 38.5 SV + + -
barcode03 3 51.6 flox_deletion - - -

Details.pdf

The output directory contains a figure of whole-allelic profile.
This is an example result of three samples.

The barcode01 is a wild-type mouse as a control, whereas the barcode02 and barcode03 are genome-edited founder mice with a flox knock-in design.

This result shows that most Nanopore reads of barcode02 are labeled as "intact target" (flox), and indicates that barcode02 is a candidate for the desired homozygous mice.

Consensus

The Conseusus folder includes FASTA and HTML files that display the consensus sequence in each allele.

Here is an example of DAJIN's consensus sequence using the point mutation.

BAM

The BAM folder includes BAM files from all and each allele.

The BAM files can be visualized by IGV.

License

This project is under the MIT License - see the LICENSE file for details

Citation

PLOS BIOLOGY

@article{Kuno_2022,
	title={DAJIN enables multiplex genotyping to simultaneously validate intended and unintended target genome editing outcomes},
	volume={20},
	url={https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3001507},
	DOI={10.1371/journal.pbio.3001507},
	number={1},
	journal={PLOS Biology},
	author={Kuno, Akihiro and Ikeda, Yoshihisa and Ayabe, Shinya and Kato, Kanako and Sakamoto, Kotaro and Suzuki, Sayaka R. and Morimoto, Kento and Wakimoto, Arata and Mikami, Natsuki and Ishida, Miyuki and et al.},
	year={2022},
	month={Jan},
	pages={e3001507}
}

‌