Alignment clustering to reveal functionality in biosequences: WeMine™

Description of the invention

Waterloo researchers have developed a novel software tool that discovers, locates, aligns and clusters non-redundant statistically significant sequence patterns from a diverse range of biosequence data (e.g. protein, DNA and amino acid sequences). When these patterns are integrated into a tool called WeMine, researchers will be able to discover and identify conserved sequence patterns, protein binding sites, partition subgroups and uncover co-occurring interacting regions of biomolecules. Identifying highly conserved and non-redundant sequence patterns can lead to the discovery of genetic and/or protein markers (biomarkers), which can aid in drug development, disease treatments, and uncovering crucial information contained within big data.

Advantages

• Does not require any prior knowledge to generate and analyse the data thereby reducing wet lab biology and bioinformatics costs
• Comprehensive representation of the sequence residue associations and distant structural and functional associations
• Unaffected by mislabelled or unbalanced class labelling, which is a problem for other traditional algorithms, such as Hidden Markov Model and Support Vector Machine
• The software is flexible; it can process DNA sequences, Protein sequences, and discovery patterns of DNA-Protein binding regions
• Compared to existing methods, this software is 616x faster, more accurate, and not restricted by input data parameters, than traditional motif-finding software.

Potential applications

• Drug development
• Genomic, transcriptomic and proteomic data analysis from such things as Microarray, RNA-Seq, and ChIP sequencing.
• Genetic testing/diagnostics – personalized medicine
• Biologics