OGTFinder is a tool that predicts the optimal growth temperature (OGT) for a user-provided (partial) proteome based on mean amino acid descriptors and taxonomic domain. For more details, please refer to:
Colette, S., François, J., De Moor, B., & Van Noort, V. (2026). OGTFinder: A Curated Growth Temperature Data Set and Its Application To Predict Optimal Growth Temperatures of Bacteria and Archaea. link
Installation
System Requirements: Python (>= 3.9)
Installation with pip
pip install OGTFinder
Installation with conda
conda install -c bioconda ogtfinder
Usage
Input Data
The input is a FASTA file containing proteome sequences of the organism of interest. Comment lines starting with ‘!’, ‘#’, or ‘;’ are allowed. Besides the proteome, the user also needs to specify the taxonomic domain of the organism of interest, either Archaea or Bacteria (default) with --domain.
Output Data
The prediction is detailed in the file results.tsv in the output directory, which may be specified with --outdir. The result file is a tab-delimited file consisting of the following 4 columns:
filename: Input FASTA file
domain: User-specified taxonomic domain
prediction [°C]: OGT prediction
class: Corresponding thermophilicity class
In addition, Debug mode --debug outputs a file descriptors.tsv containing the feature values inputted to the model.
Fervidobacterium pennivorans is an anaerobe, thermophilic bacterium from the phylum Thermotogota. Its optimal growth temperature is 70°C according to the database ThermoBase and the supplementary information from Lyubetsky et al. (2020), and 80°C according to the database Tempura. In this work, we consider the median OGT value as the true OGT, which in this case is 70°C. Starting from the genome GCF_000235405.2, the proteome can be predicted with a genome annotation tool such as Prokka or Bakta. Alternatively, the proteome can be downloaded from NCBI or UniProt if available. In this work, we used Prokka to annotate all genomes with the following command:
The resulting output directory Prokka contains the predicted proteome in the output file GCF_000235405.2.faa. This output file is also provided here under the tests directory.
Now, the OGT can be predicted with the following command:
OGTFinder: optimal growth temperature prediction for prokaryotes
Table of contents
3.1 Input Data
3.2 Output Data
3.3 Usage Command
3.4 Examples
Introduction
OGTFinder is a tool that predicts the optimal growth temperature (OGT) for a user-provided (partial) proteome based on mean amino acid descriptors and taxonomic domain. For more details, please refer to:
Colette, S., François, J., De Moor, B., & Van Noort, V. (2026). OGTFinder: A Curated Growth Temperature Data Set and Its Application To Predict Optimal Growth Temperatures of Bacteria and Archaea. link
Installation
System Requirements: Python (>= 3.9)
Installation with pip
Installation with conda
Usage
Input Data
The input is a FASTA file containing proteome sequences of the organism of interest. Comment lines starting with ‘!’, ‘#’, or ‘;’ are allowed. Besides the proteome, the user also needs to specify the taxonomic domain of the organism of interest, either Archaea or Bacteria (default) with
--domain.Output Data
The prediction is detailed in the file
results.tsvin the output directory, which may be specified with--outdir. The result file is a tab-delimited file consisting of the following 4 columns:filename: Input FASTA filedomain: User-specified taxonomic domainprediction [°C]: OGT predictionclass: Corresponding thermophilicity classIn addition, Debug mode
--debugoutputs a filedescriptors.tsvcontaining the feature values inputted to the model.Usage command
Examples
Example 1: Fervidobacterium pennivorans
Fervidobacterium pennivorans is an anaerobe, thermophilic bacterium from the phylum Thermotogota. Its optimal growth temperature is 70°C according to the database ThermoBase and the supplementary information from Lyubetsky et al. (2020), and 80°C according to the database Tempura. In this work, we consider the median OGT value as the true OGT, which in this case is 70°C. Starting from the genome
GCF_000235405.2, the proteome can be predicted with a genome annotation tool such as Prokka or Bakta. Alternatively, the proteome can be downloaded from NCBI or UniProt if available. In this work, we used Prokka to annotate all genomes with the following command:The resulting output directory
Prokkacontains the predicted proteome in the output fileGCF_000235405.2.faa. This output file is also provided here under thetestsdirectory.Now, the OGT can be predicted with the following command:
The predicted OGT is both printed to the screen and stored in
my_outdir/results.tsv. The predicted OGT is 65.6°C, close to the median OGT of 70°C.