Clone the repository and enter it:
git clone https://github.com/rvaldes/IPHOMED.git
cd IPHOMED
Create a conda environment with the included conda environment file.
conda env create -n iphomed -f iphomed.yml
Modify the config.yml file to include all the databases required for the analysis, including:
- Bowtie databases (Human and mouse databases)
- Pre-built databases are available at https://benlangmead.github.io/aws-indexes/bowtie
- Kraken/Bracken microbiome database
- Pre-built databases are available at https://ccb.jhu.edu/software/kraken2/index.shtml?t=downloads
- It is highly recommended to include human and mouse genomes in the database.
- Host proteomes (Human and mouse fasta files)
- Reference proteomes are available at https://www.uniprot.org/proteomes/
- cRAP protein sequences (fasta format)
- Available at https://www.thegpm.org/crap/
- Dietary proteome (refined IPHOMED database)
- IPHOMED installation folder
- Metamorpheus (CMD.dll) path.
- Locate it in your IPHOMED conda environment folder.
Memory, threads and queues of all rules can be modified in the Snakefile to optimize the resources of your server.
Copy the Snakefile and config.yml in your working directory.
Modify the config.yml file according to the details of your analysis:
- PAIRED: True/False. Single/paired-end sequencing configuration.
- HOST: HUMAN/MOUSE. Define the host in your experiment.
- SUB_DEPTH: subsampling depth for shotgun metagenomics analysis
The working directory must have the following structure:
|- Snakefile
|- config.yml
|- samples.txt
|- FASTQs/
|- *_R[12]_001.fastq.gz
|- mzML/
|- *mzML
|- ExperimentalDesign.tsv
Create the samples.txt, including the sample names "[Sample]_R1_001.fastq.gz" of all FASTQ files in FASTQs folder (shotgun metagenomics data)
| SampleA |
| SampleB |
| ... |
Create the ExperimentalDesign.tsv (in the mzML folder), including the information on the proteomics data. Example:
| FileName | Condition | Biorep | Fraction | Techrep |
|---|---|---|---|---|
| SampleA.mzML | Group_A | 1 | 1 | 1 |
| SampleB.mzML | Group_A | 2 | 1 | 1 |
| SampleC.mzML | Group_B | 1 | 1 | 1 |
| SampleD.mzML | Group_B | 2 | 1 | 1 |
| ... | ... | ... | ... | ... |
Activate the IPHOMED conda environment:
conda activate iphomed
Finally, run the command (adjust the parameters according to your cluster):
snakemake
The output iphomed folder will contain the following files:
- host.proteins.tsv: host proteins detected by IPHOMED.
- bacteria.proteins.tsv: bacterial proteins detected by IPHOMED.
- diet.proteins.tsv: dietary proteins detected by IPHOMED without quality filtering.
- diet.proteins.filtered.tsv: high-quality dietary proteins detected after IPHOMED refinement.
- diet.proteins.filtered.bySample.tsv: high-quality dietary proteins detected in each sample independently.