SYSTEM AND METHOD FOR SCORING PEPTIDE MATCHES

Patent №

US 7,409,296

Granted

2008-08-05

Filed 2003

Owner

GENEVA BIOINFORMATICS (GENEBIO) S.A.

Lab

AI components

2

ml · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

10624531

The present invention relates to a system and method for scoring peptide matches. Embodiments of the present invention improves identification of peptides and proteins by introducing an appropriate signal detection based scoring system and what is believed to be the new concept of an extended match. To score a match between a first peptide and a second peptide, a stochastic model may be generated based on one or more match characteristics associated with the first peptide, the second peptide and their fragments. A first probability that the first peptide matches the second peptide, and second probability that the first peptide does not match the second peptide, may be calculated based on the stochastic model. And a match between the first peptide and the second peptide may be scored based at least in part on a ratio between the first probability and the second probability.

Machine learningAI hardwareG01N 33/6818G16B 20/00Y10T 436/24

AI classification

AI hardware0.70
Machine learning0.60
Natural language0.35
Knowledge representation0.03
Planning0.00
Speech0.00
Evolutionary computation0.00
Vision0.00

Ownership

GENEVA BIOINFORMATICS (GENEBIO) S.A.

assignment · 148320317

Assignors

COLINGE, JACQUES, MASSELOT, ALEXANDRE

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC