HighGene

HighGene is a Python package enabling the visualisation of normalised gene counts. Instead of displaying absolute gene counts, will generate a figure with normalised counts based on the total amount of genes per species. Subsequently the Z-Score is calculated for each class. HighGene contains analyses built specifically for JGI's Mycocosm formatted Gene Class Classification results (eg., KOG/KEGG Classes), but allows for the visualisation of custom files.

Installation

HighGene can be installed by first cloning this GitHub repository. Subsequently, go into the location where the repository was saved and use pip to install the package.

# Clone repository
git clone https://github.com/WesterdijkInstitute/HighGene.git

# Go into repository location
cd .../HighGene/

# Install package with pip
pip install .

Usage

HighGene requires 3 inputs, the input folder, output folder and file name and the analysis mode.

usage: HighGene [-h] -i INPUT -o OUTPUT -m {0,1,2,3,4,5}

HighGene - Highlighting gene content varation, For more information see: https://github.com/WesterdijkInstitute/HighGene

options:
  -h, --help            show this help message and exit
  -i INPUT, --input INPUT
                        Input folder
  -o OUTPUT, --output OUTPUT
                        Output folder + File name
  -m {0,1,2,3,4,5}, --mode {0,1,2,3,4,5}
                        Analysis modes:
                            0: Custom File
                            1: Transcription Factors
                            2: Peptidase Clans
                            3: CAZymes
                            4: KEGG
                            5: KOG

In addition to a protein fasta file of the proteome of each species, the different analysis modes all require different files to be present in the input folder. They are the following:

• Custom file:

  • A .tsv file with two columns and headers for each species.
    • 1: Protein ID
    • 2: Class prediction

• Transcription Factors:

  • An InterPro classification file (JGI) for each species. 'IPR' must be included in the file name
  • A .tsv file named 'TF_PFAM.tsv' containing the Pfam Accessions for the transcription factors of interest.

• Peptidase Clans:

  • An InterPro classification file (JGI) for each species. 'IPR' must be included in the file name
  • A .tsv file named 'Pep_PFAM.tsv' containing the Pfam Accessions for the peptidase clans of interest.

• CAZymes:

  • A CAZyme prediction results from dbCAN3 for each species. 'CAZy' must be included in the file name

• KEGG:

  • A KEGG classification file (JGI) for each species. 'KEGG' must be included in the file name

• KOG:

  • A KOG classification file (JGI) for each species. 'KOG' must be included in the file name

Important formatting information

• Headers of classification files should always start with a #

• It is important that all files start with the same abbreviated species name that is also used in the protein fasta file followed by an underscore (e.g., Species1_Genes.fasta). Additionally, the abbreviated species names may NOT contain an underscore (e.g., Species1_A_Genes.fasta). If this is the case in the protein fasta files, change the names accordingly. HighGene uses this to determine which fasta file belongs to which gene classification files.

• When downloading protein fasta files from JGI's Mycocosm, the fasta headers will be formatted as: 'jgi|Species1|0000001|...'. HighGene requires this fasta header format. If a different header is present, change it to something that looks like JGI's format (e.g., abc|Species1).

• Examples of all file format types can be found under the 'Examples' folder.

Author

Tim Verschuren
GitHub LinkedIn