http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-109300503-B
Outgoing Links
| Predicate | Object |
|---|---|
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G16B15-20 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G16B40-00 |
| filingDate | 2018-08-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| grantDate | 2021-08-03-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationDate | 2021-08-03-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-109300503-B |
| titleOfInvention | Global and local lower bound estimation synergistic group protein structure prediction method |
| abstract | A global and local lower bound estimation collaborative population protein structure prediction method is characterized in that under the framework of a differential evolution algorithm, firstly, a global lower bound estimation model of an energy model is established by calculating a support vector of each conformational individual in a current population; then, aiming at each test conformation, selecting a part of conformations adjacent to the test conformation to construct an adjacent conformation sub-population, and then randomly selecting the conformation from the adjacent conformation to guide a variation process; secondly, calculating an energy local lower bound estimation value according to the support vector of the conformation closest to the test conformation, and calculating a corresponding energy global lower bound estimation value according to a global lower bound estimation model; and finally, selecting the lower one of the global and local energy lower bound estimation values to guide the selection process of the conformation, and reducing the evaluation of the energy function. The invention provides a global and local lower bound estimation collaborative population protein structure prediction method with low calculation cost and high prediction efficiency. |
| priorityDate | 2018-08-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 13.