.png)
BioAct.AI
DRUG DISCOVERY PLATFORM
An AI-assisted platform designed to enhance the quality and efficiency of discovering drugs derived from natural compounds
We're building a cutting-edge drug discovery platform focused on nature-based compounds
We use AI to swiftly screen plant-based chemicals with promising bioactive signals. Our platform combines extensive data libraries and proprietary machine learning models to identify potential compounds, even for the toughest targets, accelerating drug discovery and bringing new treatments to market faster and more efficiently
01. Plant Genomics/Chemistry
Generating natural compound library from tropical medicinal herb
Natural-based Small Molecules Library
We cultivate a diverse array of medicinal plants in-house to create a comprehensive library of small molecules. These naturally derived compounds are potential drug candidates, providing a rich source for novel therapeutic discoveries.
Metagenomics and Metaproteomics Analysis
By analyzing genetic material directly from environmental samples, we uncover novel bioactive compounds and explore the microbial diversity associated with medicinal plants. Our metaproteomic studies further investigate protein expressions, revealing new therapeutic proteins and enzymes.
ChEMBL Library
Utilizing the ChEMBL database, a manually curated repository of bioactive molecules with drug-like properties, we identify potential drug candidates and gain insights into their biological activities.
02. AI and Computation Biology
Shortlisting potential candidates by matching our compounds with the druggable protein library
Cross Attention Between Protein Language Models (PLM) and Moleculars Graph
Our advanced AI models predict interactions between proteins and small molecules by integrating protein language models with molecular graphs. This enhances the accuracy of identifying promising drug candidates.
Drug-Target Interaction Graph Neural Network (DTIGN)
DTIGN employs cutting-edge graph neural networks to model and predict interactions between drugs and their targets. This approach identifies the most promising drug candidates based on their interaction patterns.
Read moreMolecular Dynamic Simulations
Our molecular dynamic simulations provide detailed insights into the structural and functional dynamics of biomolecules. These simulations are crucial for drug design and optimization, ensuring our candidates are effective and safe.
03. Biomedical
Validating candidates by in vitro and in vivo studies
Preclinical Validation
We conduct thorough preclinical testing of our drug candidates in non-human subjects to evaluate their safety, efficacy, and pharmacokinetics. This essential step ensures only the most promising candidates proceed to human trials.
Mechanistic and ADMET Studies
Our mechanistic studies delve into the actions of drug candidates, while ADMET evaluations assess their Absorption, Distribution, Metabolism, Excretion, and Toxicity properties. These studies ensure our candidates are viable for clinical use.
01. Plant Genomics/Chemistry
Generating natural compound library from tropical medicinal herb
Natural-based Small Molecules Library
We cultivate a diverse array of medicinal plants in-house to create a comprehensive library of small molecules. These naturally derived compounds are potential drug candidates, providing a rich source for novel therapeutic discoveries.
Metagenomics and Metaproteomics Analysis
By analyzing genetic material directly from environmental samples, we uncover novel bioactive compounds and explore the microbial diversity associated with medicinal plants. Our metaproteomic studies further investigate protein expressions, revealing new therapeutic proteins and enzymes.
ChEMBL Library
Utilizing the ChEMBL database, a manually curated repository of bioactive molecules with drug-like properties, we identify potential drug candidates and gain insights into their biological activities.
02. AI and Computation Biology
Shortlisting potential candidates by matching our compounds with the druggable protein library
Cross Attention Between Protein Language Models (PLM) and Moleculars Graph
Our advanced AI models predict interactions between proteins and small molecules by integrating protein language models with molecular graphs. This enhances the accuracy of identifying promising drug candidates.
Drug-Target Interaction Graph Neural Network (DTIGN)
DTIGN employs cutting-edge graph neural networks to model and predict interactions between drugs and their targets. This approach identifies the most promising drug candidates based on their interaction patterns.
Molecular Dynamic Simulations
Our molecular dynamic simulations provide detailed insights into the structural and functional dynamics of biomolecules. These simulations are crucial for drug design and optimization, ensuring our candidates are effective and safe.
03. Biomedical
Validating candidates by in vitro and in vivo studies
Preclinical Validation
We conduct thorough preclinical testing of our drug candidates in non-human subjects to evaluate their safety, efficacy, and pharmacokinetics. This essential step ensures only the most promising candidates proceed to human trials.
Mechanistic and ADMET Studies
Our mechanistic studies delve into the actions of drug candidates, while ADMET evaluations assess their Absorption, Distribution, Metabolism, Excretion, and Toxicity properties. These studies ensure our candidates are viable for clinical use.
