Identifies bends from local sequence information. TMKink is a neural network predictor which identifies over two-thirds of all bends. TMKink is a prediction method that provides useful information that can be incorporated into helix packing algorithms. Kink prediction will only improve in accuracy and refinement as more structures become available.
Medical Scientist Training Program, UCLA-DOE Institute for Genomics and Proteomics, Molecular Biology Institute, UCLA, Los Angeles, CA, USA; Department of Structural Biology, Weizmann Institute of Science, Rehovot, Israel; Department of Chemistry and Biochemistry, UCLA-DOE Institute for Genomics and Proteomics, Molecular Biology Institute, UCLA, Los Angeles, CA, USA
TMKink funding source(s)
This work was supported by NIH Grant number: RO1 GM063919, Grant sponsors: Ruth L. Kirschstein NRSA Predoctoral Fellowship Award to Promote Diversity in Health-Related Research, Molecular Biology Whitecome Stipend.