Fig. 1

The PathoFact pipeline. a Framework of the PathoFact pipeline. The pipeline consists of three different modules related to (i) virulence factors, incl. (ii) bacterial toxins, and (iii) antimicrobial resistance genes. SignalP is incorporated for the prediction of secreted toxins and virulence factors. All modules can either be run independently or jointly. b Classification framework for the prediction of virulence factors. The prediction of virulence factors depends on two different aspects: (i) a HMM domain database, (ii) a random forest classifier. Sequences predicted positive from both are classified as virulence factors. The incorporation of SignalP in the framework allows integration of information regarding the likely secretion of the virulence factors