Chorus-Reproduction-Scripts

This repository is dedicated to the reproduction of the manuscript titled "Rapid Multiple Protein Sequence Search by Parallel and Heterogeneous Computation."

The official source code is at the following link

• Download Data
• Environment
• Scrips
  • Other
  • 01.Reproduction-Review1-Result
  • 02.Supplementary-Figure9-Reliability-benchmark-with-competing-methods
  • 03.Supplementary-Figure8-Performance-cross-over-point-with-competing-methods
  • 04. Cas-analysis scrips
• Contact

Download Data

• MMseq2 benchmark

cd /home/wangziyuan/fxw/Chorus-Reproduction-Scripts
wget http://wwwuser.gwdg.de/~compbiol/mmseqs2/mmseqs2-benchmark.tar.gz
tar -zxvf mmseqs2-benchmark.tar.gz

Environment

Operating System: Ubuntu 20.04.5 LTS CPU: Xeon E5-2680 v4

Software	Version
chorus	1.0.0
mmseqs2	45111b641859ed0ddd875b94d6fd1aef1a675b7e
blast	2.12.0

Set the environment variable (optional)

export CHORUS_HOME="/home/wangziyuan/bio/Chorus/bin"
export PATH=$CHORUS_HOME:$PATH

export MMSEQS_HOME="home/wangziyuan/bio/mmseqs/mmseqs/bin/"
export PATH=$MMSEQS_HOME:$PATH

export BLAST_HOME="/home/wangziyuan/bio/blast-2.12.0/ncbi-blast-2.12.0+/bin"
export PATH=$BLAST_HOME:$PATH

Remember to replace the paths in the XX_HOME variables according to your actual installation locations.

Scrips

├── mmseqs2-benchmark-pub
├── README.md
└── scrips
    ├── 01.Reproduction-Review1-Result
    │   ├── blastp.sh
    │   ├── chorus.sh
    │   └── mmseqs2.sh
    ├── 02.Supplementary-Figure9-Reliability-benchmark-with-competing-methods
    │   ├── chorus.sh
    │   ├── diamond.sh
    │   ├── evaluation
    │   └── mmseqs2.sh
    ├── 03.Supplementary-Figure8-Performance-cross-over-point-with-competing-methods
    │   ├── chorus.sh
    │   ├── diamond.sh
    │   └── mmseqs2.sh
    ├── 04.Cas-Analysis
    │   └── get.sh
    └── other
        └── drop-cache.sh

Other

In the operating system, Page Cache and Slab Cache can accelerate I/O read and write operations. Due to the acceleration effects of these system caches, there can be inconsistencies in runtimes for the same query and target. To ensure fairness in our tests, we clear the system's Page Cache and Slab Cache before each search operation.

File Name	Discription
drop-cache.sh	drop the system cache

01.Reproduction-Review1-Result

the query is the query.fasta (6370 queries) and targetdb.fasta (30430281 targets) in the mmseqs2-benchmark-pub

File Name	Discription
blastp.sh	*.sh is the script for running the corresponding program.
chorus.sh	*.sh is the script for running the corresponding program.
mmseqs2.sh	*.sh is the script for running the corresponding program.

02.Supplementary-Figure9-Reliability-benchmark-with-competing-methods

the query is the query.fasta (6370 queries) in the mmseqs2-benchmark-pub the target is the targetannotation.fasta (3374007 targets)

we use the tantan tool to tantan the targetannotation.fasta

1. Download and build tantan: https://gitlab.com/mcfrith/tantan

2. Tantan the query and database

tantan -p -x X query.fasta query_tantan.fasta
tantan -p -x X targetannotation.fasta targetannotation_tantan.fasta

3. Run the script

File Name	Discription
blastp.sh	*.sh is the script for running the corresponding program.
chorus.sh	*.sh is the script for running the corresponding program.
mmseqs2.sh	*.sh is the script for running the corresponding program.
evaluaion	This file is the result of a search analyzed with the evaluate tool of mmseqs-benchmark.

03.Supplementary-Figure8-Performance-cross-over-point-with-competing-methods

In this part, we use the script to analyze the time of mmseqs2, chorus, and diamond to get the cross-over point.

The database is UniRef90 (Mar 2023 version, size 75G).

The sequences in the querys set were randomly extracted from UniRef50 (June 2015 version), encompassing a size of 14 GB, with each sequence having a length of 1000 amino acids.

You can get the S8_CrossoverPointQuerys.zip from this repo's Release or You can use this command to download the sequences.

wget https://github.com/FanxwGit/Chorus-Reproduction-Scripts/releases/download/v1.0/S8_CrossoverPointQuerys.zip

File Name	Discription
chorus.sh	*.sh is the script for running the corresponding program.
diamond.sh	*.sh is the script for running the corresponding program.
mmseqs2.sh	*.sh is the script for running the corresponding program.
S8_CrossoverPointQuerys.zip	a sequences randomly extracted

04. Cas-analysis scrips

Because the cas-analysis scipts is too large, we put it in this repository's release. You can download it from this repo's Release

or you can use this command to download it.

wget https://github.com/FanxwGit/Chorus-Reproduction-Scripts/releases/download/v1.0/crispr_query.zip
unzip crispr_query.zip

More details about Cas-analysis in this notes.md

after the extraction is completed, the expected folder structure is

/home/username
├── casq-demo.casq
│   ├── casq-known_CasNuevo
│   ├── casq.p1
│   ├── casq.p1.log
│   ├── casq.p1start
│   ├── notes
│   ├── run.initOrf.log
│   └── run.initOrf.sh
└── cast-build
    ├── casq-known_CasNuevo
    ├── CAST
    └── cast-knowncas
└── notes.md

cast-build is the program/script folder casq-demo is the demostration project folder notes.md is the notes for the cas-analysis

File Name	Discription
crispr_query.zip	This Zip contain the cas-analysis scrips.

Contact

Thank you for your interest in our work. If you have any questions, please commit an issue.