Biomaterial Database

Learning to discriminate between ligand-bound and disulfide-bound cysteines. 2004

Andrea Passerini, and Paolo Frasconi

Dipartimento di Sistemi e Informatica, Università a di Firenze, 50139 Firenze, Italy. passerini@dsi.unifi.it

We present a machine learning method to discriminate between cysteines involved in ligand binding and cysteines forming disulfide bridges. Our method uses a window of multiple alignment profiles to represent each instance and support vector machines with a polynomial kernel as the learning algorithm. We also report results obtained with two new kernel functions based on similarity matrices. Experimental results indicate that binding type can be predicted at significantly higher accuracy than using PROSITE patterns.

UI	MeSH Term	Description	Entries
D008024	Ligands	A molecule that binds to another molecule, used especially to refer to a small molecule that binds specifically to a larger molecule, e.g., an antigen binding to an antibody, a hormone or neurotransmitter binding to a receptor, or a substrate or allosteric effector binding to an enzyme. Ligands are also molecules that donate or accept a pair of electrons to form a coordinate covalent bond with the central metal atom of a coordination complex. (From Dorland, 27th ed)	Ligand
D003545	Cysteine	A thiol-containing non-essential amino acid that is oxidized to form CYSTINE.	Cysteine Hydrochloride,Half-Cystine,L-Cysteine,Zinc Cysteinate,Half Cystine,L Cysteine
D004220	Disulfides	Chemical groups containing the covalent disulfide bonds -S-S-. The sulfur atoms can be bound to inorganic or organic moieties.	Disulfide
D000465	Algorithms	A procedure consisting of a sequence of algebraic formulas and/or logical steps to calculate or determine a given task.	Algorithm