Promotheus is a data-driven pipeline designed to retrieve, process, and integrate publicly available transcriptomic data from bacterial species and their strains. Its main goal is to support the selection of promoters that exhibit either stable (constitutive) or variable (inducible) expression profiles, facilitating biosensor design in synthetic biology applications. Promotheus performs data retrieval from public repositories, raw data pre-processing, normalization, and batch effect correction. Gene IDs harmonization is performed through orthogroup identification, allowing a multi-strain and/or multi-species comparisons.

To learn more about it, read the manuscript at XXX.

Promotheus is deployed via Docker. If you do not already have Docker installed, follow the instructions for docker installation at https://docs.docker.com/

Once Docker is installed, you can pull the Promotheus image from Docker Hub by running:

docker pull ste40/promotheus

• ASMcode must be specified. If not provided, the script will terminate. This genome is not used for reads mapping, but rather for orthology-based mapping of GeneIDs across species, reported in the final output table.
• If OPERONS_THRESHOLD is greater than the number of GSEs available for a given specie, it will be adjusted to match the number of available GSEs.
• When MULTITAXA = TRUE, ComBat normalization is performed separately for each taxon, assigning the GSEs of each taxon to distinct batches.

The pipeline needs two separate input tables for RNA-Seq and Microarray datasets. Each table must contain the following mandatory columns:

• Accession: Accession codes for various datasets (GSEs).
• taxon: Microorganism names, based on NCBI taxonomy name.
• reference_genome: Assembly accession codes for the reference genome used for reads alignment and gene counts quantification.

The examples of these two tables (RNA-Seq and Microarray) are provided in this repository, see Examples directory.

NOTE: It is reccomended to use our GEO_Search.R script to generate these two tables. See below for more information.

We provide GEO_Search.R script to help user to find RNA-Seq and Microarray experiments on NCBI GEO Datasets based on a provided taxonomic ID (taxID).

Features:

• Taxonomy Retrieval: Retrieves the scientific name and subspecies for a given taxID.
• Genome Assembly: Fetches the reference genome assembly version for the organism.
• GEO Datasets Search: Searches for RNA-Seq and Microarray datasets from GEO, filtering by dataset type and minimum sample count (4).
• Export: Outputs filtered results to Excel files for RNA-Seq and Microarray experiments.

R (version 3.6 or higher recommended) must be installed in your system.

• Required R packages:
  • devtools
  • rentrez
  • openxlsx
  • optparse

To install the required packages, run the following in R:

install.packages(c("devtools", "openxlsx", "optparse"))
devtools::install_github("ropensci/rentrez")

Then, the script can be run from command-line, specifying the taxID of interest:

Rscript GEO_Search.R --taxID <taxID>

NOTE: The script attempts to retrieve data from NCBI, so it may fail if the servers are down or experiencing issues.

The pipeline runs within a Docker container. You can set environment variables using the -e flag. Make sure to mount your input/output directories with the -v flag.

WARNING: the scripts provided in this repository can't be executed without the appropriate environment and softwares/packages, which are included in Docker's image.

Below is an example command to run the Promotheus pipeline. Adjust the paths and parameters as needed:

  -e THREADS=4 \
  -e MEMORY=4000 \
  -e CORRECTION_DE=BH \
  -e PVALUE=0.01 \
  -e OPERONS_DETECTION=TRUE \
  -e AMINO_DIFF=5 \
  -e MULTITAXA=TRUE \
  -e SIMILARITY=30 \
  -e ASMcode=ASM317683v1 \
  -e OPERONS_THRESHOLD=3 \
  -v /path/to/input/:/input \
  -v /path/to/output/:/output \
  ste40/promotheus

The output table contains includes gene expression values, percentiles, coefficient of variation (CV), and Differential expression analysis results.

• Gene_name: The name of the gene.
• GeneID: The unique identifier for the gene, according to the gene nomenclature of the user-provided ASMcode.
• Expression_GSEXXXXX: Gene expression values for each GSE.
• Mean_Expression: The average expression value of the gene across all samples in the datasets.
• CV: Coefficient of variation, indicating the variability of the gene expression across samples, computed as the standard deviation of log-transformed expression values. A higher CV suggests more variation in gene expression.
• Percentile_GSEXXXXX: Percentile values for the gene expression in different datasets.
• Mean_Percentile: The average percentile value across all datasets for the gene.
• Percentile_SD: Standard deviation of the percentile values across all datasets for the gene.
• DE: Differential expression status (YES|NO), in at least one GSE.
• Sequence: Contains the DNA sequence upstream of the gene.

An example of output table is provided in this repository (Output.xlsx)