Business
IPA’s Subsidiary, BioStrand, Provides an Update on LENSai™
BioStrand Unveils Groundbreaking Retrieval Augmented Generation (RAG)-Based LLM Platform Integrated with Patented HYFT Technology BioStrand's Work
About this update from Mindwalk Holdings Corp.
[{"type":"text","content":"\nBioStrand Unveils Groundbreaking Retrieval Augmented Generation (RAG)-Based LLM Platform Integrated with Patented HYFT Technology\n\n\nBioStrand's Work Exemplifies the IPA Family's Investment in Advanced AI, Aiding Partners in Developing New Biologics for Previously Undruggable Targets\n\n\n VICTORIA, British Columbia--(BUSINESS WIRE)--\nImmunoPrecise Antibodies Ltd. (NASDAQ: IPA) (“ImmunoPrecise” or “IPA” or the “Company”), an AI-driven biotherapeutic research and technology company, announces that its subsidiary, BioStrand®, has commercially launched its state-of-the-art Retrieval Augmented Generation (RAG)-based Large Language Model (LLM) platform. This pioneering platform seamlessly integrates with the Company’s patented HYFT technology and LENSai platform, signifying a noteworthy leap in the market as the Company aims at ensuring unparalleled accuracy, interpretability, and data-centric design in generative AI tools.\n\n\nBioStrand's innovative approach to solving the Information Integration Dilemma (IID) has led to the development of a unique technology design that encapsulates and unifies diverse data modalities. This includes syntactical (sequence) data, 3D structural data, unstructured scientific information (e.g., scientific literature), all integrated within a singular framework, the LENSai Knowledge Graph. This breakthrough facilitates efficient data fusion, enabling a comprehensive analysis and interpretation of complex biological data.\n\n\nKnowledge Graphs and LLMs have been recognized for their superior performance over conventional approaches in drug discovery. BioStrand’s integration of their proprietary and patented technologies with LLM synergizes strengths and addresses limitations, leading to a more efficient drug discovery platform. Biomedical LLMs, specifically those pre-trained on domain-specific vocabulary, outperform traditional tools in many biological data-based tasks. For instance, for the important step of identifying drug targets, AI-powered language models have demonstrated superiority over even the most state-of-the-art approaches. Furthermore, AI-enabled LLMs are now being utilized across the drug discovery and development pipeline for predicting drug-target interactions, molecular properties, and even potential drug withdrawals due to safety concerns.\n\n\n\n\nKey Features of BioStrand’s LEN...