Supplement to LIBRUS: Combined Machine Learning and Homology
Information for Sequence-based Ligand-Binding Residue Prediction.
This page contains data supplementary to the article submitted to Bioinformatics
2009. Below are a description of the files.
- cv.folds
- List of protein chains used in cross validation and their
fold. Referred to as DS1 in the paper.
- findsite.folds
- List of a subset of protein chains from the benchmark used by
FINDSITE, used to compare LIBRUS to FINDSITE. Referred to as DS2 in
the paper.
The following programs were developed to aid the study. They are
provided in binary format for the Linux x86 architecture
(32-bit). Some help is given by passing the -help option
to each program. If additional aid is required, please e-mail the
authors.
- pdbtool
- Program to extract sequence and ligand information from PDB files.
- aligntool
- Program to align two sequences, allows various input formats and
scoring mechanisms.
- seqdbtool
- Program to create a database of sequences from various features
(PSSMs, SSEs, etc).
- align_db
- Program to allow the alignment of queries against a database of sequences.
- align_countcols
- Program to count aligned columns of ligands according to the
results of align_db