Identifies RNA binding sites and characterizes DNA-binding protein specificity. DeepBind utilizes a set of sequences and an experimentally determined binding score for each sequence. This software is built on deep learning, a scalable and modular pattern discovery method and doesn’t need common application-specific heuristics like seed finding. It can discover new patterns even when the locations of patterns within sequences are not known.
Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON, Canada; Donnelly Centre for Cellular and Biomolecular Research, University of Toronto, Toronto, ON, Canada; Canadian Institute for Advanced Research, Programs on Genetic Networks and Neural Computation, Toronto, ON, Canada; Center for Autoimmune Genomics and Etiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA; Divisions of Biomedical Informatics and Developmental Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA
DeepBind funding source(s)
Supported by a grant from the Canadian Institutes of Health Research (OGP-106690), a John C. Polanyi Fellowship Grant, the Canadian Institutes for Advanced Research, a joint Autism Research Training and NeuroDevNet Fellowship, and a Fellowship from the Natural Science and Engineering Research Council of Canada.