Machine learning methods can replace 3D profile method in classification of amyloidogenic hexapeptides
- Equal contributors
1 Institute of Computer Engineering, Control and Robotics, Wroclaw University of Technology, 50-370, Wroclaw, Poland
2 Institute of Biomedical Engineering and Instrumentation, Wroclaw University of Technology, 50-370, Wroclaw, Poland
BMC Bioinformatics 2013, 14:21 doi:10.1186/1471-2105-14-21Published: 17 January 2013
Additional file 1:
Dataset of hexapeptides with calculated energies and amylogenic classification.
Format: XLSX Size: 295KB Download file
Additional file 2:
Position specific aminoacid frequencies of the training and test datasets.
Format: PDF Size: 277KB Download file
This file can be viewed with: Adobe Acrobat Reader
Additional file 3:
Detailed results of the best machine learning methods trained on the full and reduced training sets.
Format: TXT Size: 825KB Download file
Additional file 4:
Rules of ADTree 250 methods for full and reduced training sets. Columns contain the rules corresponding to each position in amyloidogenic hexapeptides in decreasing order of their importance. Yellow cells denote positive rules, purple – negative rules.
Format: XLSX Size: 16KB Download file
Additional file 5:
Amylogenic classification of our dataset obtained with reduced 3D profile (additional file 1: trainset(+) and trainset(-)) with different methods: 3D profile, FoldAmyloid and Waltz. The file includes the spreadsheets labeled according to the name of the external method.
Format: XLSX Size: 1.7MB Download file