Biological Entity Identification From Context-Rich Biomedical Text
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current knowledge graph-based methods for identifying biological targets in drug discovery are limited by the loss of contextual information during data translation and the inability to query complex biological contexts, leading to inefficiencies and high failure rates in target identification.
Innovation Solution
A machine learning model is trained on entity-linked text sequences to predict biological entity identifiers directly from free text, allowing for nuanced queries and leveraging rich contextual information, enabling the identification of multiple relevant targets with reduced human input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If knowledge graph-based methods are used to identify biological targets, then the structured data can be represented at-scale, but contextual information is lost during data translation
Solution Approach 1:
The patent segments the knowledge representation into two distinct components: a knowledge graph for structured entity relationships and a language model for contextual understanding. This segmentation allows each component to handle its strengths independently - the graph handles scale and structure while the language model preserves contextual nuances from the original text.
Solution Approach 2:
The patent introduces an intermediary language model that bridges the knowledge graph and the original text data. This intermediary processes the structured graph data through natural language processing, generating text sequences that retain contextual information while being grounded in the structured knowledge graph relationships.
2Measurement precision
If traditional manual analysis is used to assess biological target characteristics, then multiple properties can be considered, but the process is time consuming and creates delays
Solution Approach 1:
The patent replaces the manual mechanical analysis process with an automated computational system. The language model automatically evaluates multiple biological target properties by processing text sequences through machine learning, substituting human scientists' manual review with algorithmic analysis that maintains comprehensive property assessment while dramatically reducing time requirements.
3Productivity
If knowledge graph query methods are used, then target prediction can be performed, but complex biological contexts cannot be queried
Solution Approach 1:
The patent introduces dynamic adaptability by allowing the system to handle varying levels of query complexity. The language model can process simple entity queries as well as complex contextual queries with multiple conditions and relationships. The system dynamically adjusts its processing based on the input query structure, maintaining versatility while preserving productive target prediction capabilities.
Data Source
AI summary
A computer-implemented method of training a machine learning model to identify biological entities for drug discovery is disclosed. The method comprises providing a training data set comprising a plurality of entity-linked text sequences, each text sequence including a mention of a biological entity, where the biological entity is linked to a corresponding biological entity identifier from a set of possible biological entity identifiers; masking the mention of the biological entity within each text sequence; encoding each masked text sequence into an input representation for a machine learning model; and training a machine learning model to predict the unique entity identifier of the masked biological entity based on the input representation. The described method is able to utilise the full breadth of the rich contextual information available in the biomedical text corpus to predict new biological targets for drug discovery and avoids the restrictions intrinsic to relationship prediction using knowledge graphs. The ability to identify more promising, biologically relevant targets in an automated manner, significantly reduces the requirement of human input and reduces the failure rate in targets that are progressed in the drug delivery pipeline.