The CNN-T4SE is a tool that can be run in windows and linux operating systems. It is used to identify effector proteins through the amino acid sequence, evolutionary information in the form of a position-specific scoring matrix, secondary structure and solvent accessibility information of the proteins. The following steps will help you to use CNN-T4SE to identify effectors.

STEP-1: Prepare the input files of different prediction methods

PSSSA: To use PSSSA Encoding, the studied amino acid sequence(s) should be transferred to the sequences of "secondary structure" and "solvent accessibility" using the protein structure prediction tool SCRATCH , a local version of SCRATCH is also provided HERE

Onehot: The one-hot encoding procedure is integrated in the Software, just use the FASTA format sequence file as the input

PSSM: The PSSM files could be generated through a web tool POSSUM

An sample data of those three kinds of input files can be download HERE

STEP-2: Annotate Novel T4SE Protein(s) Using CNN-T4SE

Download the software tool CNNT4SE (for Windows)

Download the software tool CNNT4SE (for Linux)

Download the User Manual

Click to download

Click to download

Click to download

How to Cite CNN-T4SE:

J. J. Hong, Y. C. Luo, M. J. Mou, J. B. Fu, Y. Zhang, W. W. Xue, T. Xie, L. Tao*, Y. Lou*, F. Zhu*. Convolutional neural network-based annotation of bacterial type IV secretion system effectors with enhanced accuracy and reduced false discovery. Brief Bioinform. doi: 10.1093/bib/bbz120 (2019).

If you find any bug, please kindly report it to Mr. Hong ( or Prof. Zhu ( Thanks a million for using and improving CNN-T4SE, and welcome to visit our lab at

IDRB: Innovative Drug Research and Bioinformatics Group

All rights are reserved by: Innovative Drug Research and Bioinformatics Group (IDRB)
College of Pharmaceutical Sciences, Zhejiang University
Hangzhou, P.R. China, 310058.
Contact number: (86 - 571)88208444

Last Update: