System und Verfahren zur Entdeckung von Vorhergesagten Ortsspezifischen Proteinphosphorylierungskandidaten
Anmelder: FUJITSU LIMITED 🇯🇵
Details
- Veröffentlichungs-Nr.
- EP3404568
- Aktenzeichen
- EP18173025
- Anmeldetag
- 17. Mai 2018
- Veröffentlichung
- 12. Juli 2023
- Erteilung
- 12. Juli 2023
- Rechtsraum
- EP
- IPC
- G16B40/00G16B50/00G16B40/20G16B50/10G16B50/20G16B50/30G16H10/00G16B99/00
Abstract
A method for discovery of predicted site-specific protein phosphorylation statements, the statements linking a kinase, a predicted phosphorylation site, and a protein substrate, the method comprising: using data from a phosphorylation data set storing known site-specific statements each in a site-specific format linking a kinase, a phosphorylation site and a protein substrate; a kinase family database; and a protein sequence database to produce motif-based phosphorylation statements each in a generalised format linking a kinase, a motif in the form of a generalised scoring matrix for amino acids at a phosphorylation site and to either side of the phosphorylation site; and a protein substrate; generating negative statements from the motif-based statements; generating candidate statements in the generalised format by combining the motif-based statements; generating negative candidate statements by combination of the candidate statements with the negative statements; using the candidate statements and negative candidate statements in a statistical relational learning model to score the candidate statements and retain predicted motif-based phosphorylation statements in the generalised format; receiving data inputs from the protein sequence database and the predicted motif-based phosphorylation statements in the generalised format and producing predicted site-specific phosphorylation statements in the site-specific format.
Anmelder
- Firma
- FUJITSU LIMITED
- Land
- 🇯🇵 Japan
Japanisches Unternehmen für Informationstechnik, das Computer, Server, Netzwerktechnik sowie IT-Dienstleistungen und Softwarelösungen entwickelt und anbietet.
9.047 Patente in unserer Datenbank
Fachgebiete
Vertreten von
-
Haseltine Lake Kempner LLP
Haseltine Lake Kempner LLP · München
-
Haseltine Lake LLP
Haseltine Lake LLP · München