Carbon Silicon AI Announces DrugFlow 1.0 Launch and AIDD Expert Symposium

From AlphaFold 2 to ChatGPT, and then to GPT-4, artificial intelligence continues to surprise us.

So, what can AI bring to the field of new drug development?

Indeed, the transition from CADD to AIDD has seen both progress and challenges.

AI Drug Discovery Platform: DrugFlowComing soon!

We invite you to attend:“CarbonSilicon Intelligence: Born from Wisdom — Launch Event of DrugFlow 1.0, an AI-Powered New Drug R&D Platform, and Symposium on Progress and Challenges in AI-Driven New Drug Development”

Covering Key Stages of Drug Discovery:Activity Prediction, Druggability Prediction, Molecular Generation, AI ModelingFour Major Modules.

1. Activity Prediction Module——“AI + Physics” New Model:Building upon physically docked conformations, the software integrates RTMScore, the state-of-the-art rescoring function in the field, to facilitate optimal molecule selection. Furthermore, it pioneers the integration of the AI-driven docking algorithm CarsiDock, which has achieved unprecedented performance in mainstream docking accuracy benchmarks, surpassing 90% accuracy for the first time under a root-mean-square deviation threshold of less than 2 Å.

2. Druggability Prediction Module — Broad Coverage, High Speed, and Strong Interpretability:This module supports systematic evaluation of 17 physicochemical properties, 5 medicinal chemistry properties, 21 drug-likeness parameters, and 27 toxicity endpoints. Among these, the MGA method, based on multi-graph neural networks, is one of the fastest prediction algorithms currently available and offers interpretability features. Meanwhile, the MERT method, based on pre-trained Transformers, is currently one of the most accurate methods in the industry for ADMET prediction.

3. Molecular Generation Module – Comprehensive AIGC:It supports both ligand-based and protein pocket structure-based molecular generation methods. In addition to de novo molecular generation, it also enables molecular generation and optimization based on existing molecular fragments. This module integrates three generative algorithms: the MCMG algorithm, which is based on reinforcement learning and deep generative models; the ResGen algorithm, which performs autoregressive molecular generation and optimization based on protein pocket structures; and the RELATION algorithm, which comprehensively incorporates both ligand activity and protein pocket structures.

4. AI Modeling Module — Proprietary Data Modeling：For scenarios where users wish to build AI models based on their proprietary data, DrugFlow provides functionalities such as data preprocessing, data batching, descriptor definition, and machine learning algorithm modeling, while also offering AutoML and pre-training methods to enhance training accuracy.

Symposium on Progress and Challenges in AI-Driven New Drug Development

Scholars from Zhejiang University, Peking University, East China University of Science and Technology, Sichuan University, Central South University, the Shanghai Institute of Materia Medica (Chinese Academy of Sciences), the Institute of Materia Medica (Chinese Academy of Medical Sciences), and Guangzhou National Laboratory, together with industry experts from multinational corporations (MNCs) and biotech companies, have gathered to discuss progress, address challenges, and engage in online interaction. Come and listen to the insights from these leading experts!

Meeting Format: Online conference with virtual live streaming for easy participation! Scan the QR code below to register now!

CarbonSilicon Intelligence is a technology company focused on new drug development, positioning itself as a provider of artificial intelligence infrastructure and services in the field. We aim to deeply integrate cutting-edge life science technologies with information technologies such as artificial intelligence. Targeting new drug development, we leverage advanced AI techniques—including generative AI (AIGC), self-supervised pre-training, and reinforcement learning—while seamlessly integrating computational physics and hardware/software automation technologies. By enhancing capabilities in generating production data, managing data, and performing AI-based data modeling, we digitize and intelligentize every stage of the drug discovery process, establishing a closed-loop system of wet-lab and dry-lab data to address key challenges in new drug development.

DrugFlow is an AI-driven, one-stop platform for innovative drug discovery developed by CarbonSilicon Intelligence. Its core algorithms are independently owned intellectual property, and the platform is provided to third-party clients via SaaS or hybrid cloud deployment. DrugFlow covers key stages including target identification, hit discovery, and lead optimization. It integrates world-leading modules for target identification, activity prediction, druggability assessment, molecular generation and optimization, virtual screening, and AI modeling, enabling medicinal chemistry experts to identify potential drug-like molecules more efficiently and conveniently. DrugFlow is committed to building a software platform that covers the entire drug R&D lifecycle. By iteratively refining models with data generated from automated hardware and integrating expert knowledge into the workflow, it ultimately forms a unified, data-driven, human-AI collaborative design platform, significantly enhancing the predictability and success rate of drug development.