PyPredT6: A python-based prediction tool for identification of Type VI effector proteins

Area of Science:

• Computational biology and bioinformatics
• Microbiology and infectious diseases
• Machine learning in biological systems

Background:

• Effector proteins are critical in pathogen-host interactions, initiating infection by acting as the first line of invasion.
• The Type VI Secretion System (T6SS) and its unique secretion mechanism present a novel challenge for predicting T6 effector proteins.

Purpose of the Study:

• To design and develop a computational tool for the accurate prediction of T6 effector proteins.
• To establish the applicability and effectiveness of the developed tool in predicting T6 effector proteins in relevant bacterial species.

Main Methods:

• Developed PyPredT6, a standalone Python-based tool for predicting T6 effector proteins.
• Extracted 873 unique features from peptide and nucleotide sequences of experimentally verified effector proteins.
• Employed machine learning algorithms for in silico prediction of T6 effector proteins.

Main Results:

• Successfully designed and implemented PyPredT6, a novel tool for T6 effector protein prediction.
• Demonstrated the tool's applicability by performing in silico predictions in *Vibrio cholerae* and *Yersinia pestis*.

Conclusions:

• PyPredT6 offers a valuable computational approach for identifying T6 effector proteins.
• The tool's successful application in key bacterial pathogens highlights its potential for advancing research in pathogen-host interactions and T6SS biology.