HuntMi: an efficient and taxon-specific approach in pre-miRNA identification

Adam Gudyś¹, Michał Wojciech Szcześniak², Marek Sikora¹ and Izabela Makałowska², BMC Bioinformatics, 14:83 (click here to see the paper).

1. Institute of Computer Science, Faculty Of Automatic Control, Electronics And Computer Science, Silesian University of Technology, Gliwice, Poland

2. Laboratory of Bioinformatics, Faculty of Biology, Adam Mickiewicz University, Poznań, Poland

Contact: adam.gudys@polsl.pl, miszcz@amu.edu.pl

Cross validation datasets:

· data_sequences.zip – sequences used in the research in fasta format.

· data_features_base.zip – arff files containing base representation of examined sequences (21 features).

· data_features_extended.zip – arff files containing extended representation of examined sequences (21+7 features). This is a default representation for HuntMi software.

Comparative datasets:

· animals_miRBase18-19.zip – animal sequences newly introduced in miRBase 18-19 used in comparative experiments.

· plants_miRBase18-19.zip - plant sequences newly introduced in miRBase 18-19 used in comparative experiments.

Software (please refer to the README file for all the details):

· HuntMi.tar.gz – HuntMi package for miRNA classification. Requires Linux machine with Perl 5.10 (see README file when running on earlier Perl versions).

· ROCSelect.jar – Weka plugin with ROC-select procedure to be used for training models on custom datasets. Plugin works correctly with 3.6.x version of Weka.

· ROCSelect.model-cfg – Weka classifier configuration file with default ROCSelect parameters.

· README – readme file with all the instructions for miRNA identification as well as model training.

Main page