Insilico Medicine Unveils a Suite of New AI Products

Insilico Medicine (MA, US) — a clinical-stage generative AI-driven biotech company — has just announced a swathe of new products for researchers, as well as updates to Pharma.AI, its proprietary generative AI platform for early drug discovery.

The New Products

1) Dora, a large language model-driven AI assistant designed to streamline the process of drafting academic papers and other related documents

‍2) PandaOmics Box, an AI-powered research platform for localized biomarker discovery and indication prioritization

‍3) Precious-3 GPT, an LLM engine spanning multi-modal, multi-omics, multi species for biomedical research through omics data generation and interpretation

DORA

DORA is an intelligence research and writing assistant that integrates multiple AI agents that leverage LLM. It is designed to streamline the process of drafting academic papers and other scientific documents including grant and patent applications, internal research summaries, IND applications, etc.

It assists researchers in drafting these types of documents with proper referencing through engineered prompts, proprietary databases, and pre-designed content generation workflows.

To demonstrate the capabilities of DORA, Insilico collaborated with the University of Copenhagen (Copenhagen, Denmark) to submit a paper on medRxiv.

The paper drafted by DORA and later manually curated and extended, performs a comparative study about radiotherapy outcomes across brain tumor types, namely Glioblastoma Multiform and Low-Grade Gliomas based on radiotherapy phenotype and expression data from 32 cancer datasets.

Insilico plans to further test DORA in multiple types of document generation and launch a free trial version of the AI assistant to the public in late 2024.

PandaOmics Box

PandaOmics Box, is an industry-first AI-powered converged research solution that supports localized biological research while ensuring data confidentiality.

It integrates PandaOmics, Insilico’s generative AI-driven biological analysis engine, extensive scientific databases, and advanced hardware systems with state-of-the-art computational power and chip-level confidential computing for secure, offline, on-premise deployments.

PandaOmics Box is designed to significantly accelerate innovation in target identification, biomarker discovery, and indication prioritization by organizations with stringent data security requirements, especially hospitals and research institutions.

The system is equipped with OpenAPI framework, and allows Bioinformatics access to PandaOmics' extensive database and advanced AI algorithms. This feature provides researchers the flexibility to develop their own custom analyses, enabling them to create innovative methods that blend with their proprietary data for deeper insights.

Precious-3 GPT

Precious-3 GPT is a pre-trained transformer engine across multi-modal, multi-omics, multi-species for biomedical research.

Precious-3 GPT features a novel tokenization and training procedure that has enabled it to extract biomedical information from >2MM omics observations, as well as PubMed text and knowledge graphs.

It understands basic aging processes in multiple species and can discover novel protein targets, identify aging and disease treatments, predict biological age using methylation, expression, proteomics data, and blood tests. It also supports to discover anti-aging interventions and validate them virtually across multi-species, including mouse, rat, monkey, and human.

Insilico is validating part of the P3GPT's capabilities through Life Star 1, its proprietary automated robotics lab. As a next step, Insilico plans to publish the open source of Precious-3 GPT to encourage industry researchers to join in exploring anti-aging solutions using AI.