PredicTF: prediction of bacterial transcription factors in complex microbial communities using deep learning.

"we demonstrated the use of predictf in a previously sequenced p aeruginosa (pao1) genome a clinical isolate publicly available in ncbi (accession number nc_002516 2).; the transcriptomes of p aeruginosa pao1 mutants y71 y82 and y89 are available in ncbi (bioproject identifier prjna479711) [ ].; to test predictf in a complex microbial community we used an anaerobic ammonium oxidizing (anammox) microbial community from an anammox membrane bioreactor metagenome (lac_metag_1) (data publicly available at ncbi bioproject via accession number prjna511011) [ ].; these metatranscriptomes are publicly available at the european nucleotide archive under the accession numbers srr7091385 srr7523233 srr7523244 srr7523245 srr7091400 srr7091401 srr7091381 srr7091402 srr7091406 srr7523243 srr7523246.; genomes of the model organisms used in the tool validation step are available at the national center for biotechnology information ( https://www ncbi nlm nih gov/ ) under the accession numbers nc_000913 3 nc_000964 3 nc_011916 1 nc_021149 1 and nc_016830.; the datasets supporting the prediction of transcription factors in a clinical isolate of this article are available at national center for biotechnology information ( https://www ncbi nlm nih gov/ ) under the accession number nc_002516 2 (genome) and study accession prjna479711 (transcriptomes).; the datasets used for the prediction of transcription factors in complex microbial communities of this study are available at national center for biotechnology information ( https://www ncbi nlm nih gov/ ) under the study accession prjna511011.; the respective data sets of metatranscriptomes used are available at national center for biotechnology information ( https://www ncbi nlm nih gov/ ) under the sra numbers srr7091385 srr7523233 srr7523244 srr7523245 srr7091400 srr7091401 srr7091381 srr7091402 srr7091406 srr7523243 srr7523246 and the joint genome institute ( https://jgi doe gov/ ) under the gold analysis project identifiers gp0267156 gp0267150 gp0267154 gp0267155 gp0267157 gp0267158 gp026715 gp0267159 gp0267152 gp0267153 gp0267160."

"predictf is an open-source software available at https://github com/mdsufz/predictf .; the complete set of transcription factors in bactfdb used to train and test the deep learning models and the model itself (defined as predictf) are publicly available in ( https://github com/mdsufz/predictf ).; further description of the mapping of the transcriptomes to the genomes is available at https://github com/mdsufz/predictf .; all analysis results and scripts used to generate figures are available at https://github com/mdsufz/predictf ."

"Funding Open Access funding enabled and organized by Projekt DEAL."