Machine Learning aided Prediction of Plasmid Permsiveness

The package predicts a relative value for permissiveness from any arbitrary 16s rRNA input data, using a random forest model. Plasmid permissiveness is the ability of recipient bacteria to receive external DNA through the mechanism of conjugation. Prediction is made in the relative model, i.e. the permissivenss value is compared with other strains in the training dataset and reported as the % of strains having a smaller permissiveness in the training dataset. Furthermore, the package reports the closest systematic type, based on the selected rank, to the input sequence. The prediction is made for the broad-range plasmids of pB10, pKJK5 and RP4.

Danesh Moradigaravand, Liguan Li, Arnaud Dechesne, Joseph Nesme, Roberto de la Cruz, Huda Ahmad, Manuel Banzhaf, Søren J Sørensen, Barth F Smets, Jan-Ulrich Kreft Plasmid Permissiveness of Wastewater Microbiomes can be Predicted from 16S rRNA Sequences by Machine Learning, Bioinformatics, 2023;, btad400 https://doi.org/10.1093/bioinformatics/btad400