US 2024/0265993 A1 · METHOD FOR TRAINING VECTOR MODEL AND GENERATING NEGATIVE SAMPLE

Abstract

A method for training a vector model, including: obtaining more than one RNA sequence and more than one protein sequence; obtaining more than one first RNA vector by vectorizing the more than one RNA sequence; obtaining more than one first protein vector by vectorizing the more than one protein sequence; determining an interaction between the RNA sequence and the protein sequence according to the first RNA vector and the first protein vector; obtaining a similarity of more than one RNA-RNA pair by calculating a distance between any two RNA sequences; obtaining a similarity of more than one protein-protein pair by calculating a distance between any two protein sequences; training the vector model according to an interaction between the RNA sequence and the protein sequence, the similarity of the RNA-RNA pair and the similarity of the protein-protein pair.

Claims25

Dependent claims

A method for training a vector model, comprising:

obtaining at least one Ribonucleic acid (RNA) sequence and at least one protein sequence;
obtaining at least one first RNA vector by vectorizing the at least one RNA sequence;
obtaining at least one first protein vector by vectorizing the at least one protein sequence;
determining an interaction between the RNA sequence and the protein sequence according to the first RNA vector and the first protein vector;
obtaining a similarity of at least one RNA-RNA pair by calculating a distance between any two RNA sequences;
obtaining a similarity of at least one protein-protein pair by calculating a distance between any two protein sequences;
and
training the vector model according to the interaction between the RNA sequence and the protein sequence, the similarity of the RNA-RNA pair and the similarity of the protein-protein pair.

5 more elements

A method for generating a negative sample, comprising:

obtain a positive RNA-protein pair;
obtaining, by vectorizing a target RNA sequence and a target protein sequence in the positive RNA-protein pair through a trained vector model, a corresponding second RNA vector and second protein vector;
and
obtaining, based on the second RNA vector and the second protein vector, a target negative RNA-protein pair corresponding to the positive RNA-protein pair.

1 more element

A method for generating a negative sample, comprising:

obtaining a positive sample, wherein the positive sample comprises two biomolecular sequences;
obtaining, by vectorizing a first biomolecule sequence in the positive sample through a trained network model, a biomolecule vector of the first biomolecule sequence;
calculating a similarity between the biomolecule vector of the first biomolecule sequence and a biomolecule vector of any homologous biomolecule sequence except for the first biomolecule sequence;
determining a homologous target biomolecule sequence similar to the first biomolecule sequence according to the similarity;
and
obtaining a negative sample corresponding to the positive sample from the homologous target biomolecule sequence and a second biomolecule sequence in the positive sample.

3 more elements

Dependencies are inferred from the claim text.

Method for training vector model and generating negative sample

Life of this patent

Method for training vector model and generating negative sample

Life of this patent

Abstract

Claims25