Machine learning-guided peptide design
By using encoder and decoder networks to iteratively adjust the embedding of protein or peptide sequences, the problem of difficulty in generating or modifying protein or peptide sequences with specific functions in existing technologies is solved, and efficient and accurate function prediction and sequence generation are achieved.
CN115136246BActive Publication Date: 2025-09-09FLAGSHIP PIONEERING INNOVATIONS VI LLC
2 Cites 0 Cited by
Patent Information
- Application Number
- CN202080067045.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-08-02
- Filing Date
- 2020-07-31
- Publication Date
- 2025-09-09
- Estimated Expiration
- 2040-07-31
AI Technical Summary
Technical Problem
Existing technologies make it difficult to efficiently generate or modify protein or polypeptide sequences with specific functions or properties through functional evaluation, and traditional methods lack accuracy and reproducibility.
Method used
An engineering approach is used to generate embeddings of protein or peptide sequences through encoder and decoder networks, and the embeddings are iteratively adjusted using a supervised model and decoder network to achieve a target level of functionality, generating improved biopolymer sequences.
Benefits of technology
It achieves efficient generation and modification of protein or peptide sequences based on functional evaluation, and improves the accuracy and reproducibility of functional prediction.
✦ Generated by Eureka AI based on patent content.
Abstract
Systems, devices, software, and methods for engineering amino acid sequences configured to possess specific protein functions or properties. The methods implement machine learning to process an input seed sequence and generate an optimized sequence with the desired function or property as output.
Need to check novelty before this filing date? Find Prior Art