AI Empowers Drug Discovery

Silexon is an emerging AI-empowered technology company which aims to create an open AI platform for strategic collaboration in order to facilitate data-driven life science research and empower drug R&D process, and ultimately provide patients with greater access to innovative drugs for unmet medical needs.

MORE
About US

Silexon is an emerging AI-empowered technology company which aims to create an open AI platform for strategic collaboration in order to facilitate data-driven life science research and empower drug R&D process, and ultimately provide patients with greater access to innovative drugs for unmet medical needs.

We are the only AI-driven company that has established strategic collaobrative relationship with two leading CROs (i.e. Bioduro and Viva) with respect to AI modeling as well as co-promotion of each other's capabilities.

Silexon has reached co-development agreement with about 20 partners including but not limited to MNCs, public pharma companies, biotech companies and regulatory agency.

We have developed and have been continuously optimizing several generations of proprietary Drug-Target-Interaction models within our AI4D™ platform which leverage our inter-disciplinary capabilities and know-how.

Our AI/ML models integrate sequence-, structure-, interaction and ligand-based information in order to efficiently find hits and leads.

AI Models/
Technological Platform

AI4D™ - AI for Drug Discovery, Design and Development

Our AI/ML platform (i.e. AI4D™) is focused on early drug discovery for small molecules, with several generations of proprietary Drug-Target-Interaction models, and we have additional proprietary models for target ID, molecule generation, DMPK, toxicity/DILI, and more.

Silexon is seeking partners to work closely on FIC and difficult-to-drug targets, and also on repurposing opportunities.

Learn more
Our Vision

To provide patients with greater access to innovative drugs for unmet medical needs


Our Mission

To create an open AI platform for collaboration to facilitate data-driven biopharma research and empower drug R&D process and real-world study

People Behind Our Success

Founder

Dr. Jianyang ZENG

Founder

Associate Professor (tenure-track) at IIIS, Tsinghua Univ.
Expert in computational biology & AI/ML/DL-enabled biopharma researches
Ph.D. in Computer Science from Duke Univ.
70+ publications on CNS & other peer-reviewed periodicals.
2019 “Wu Wen Jun AI Science & Technology Award” 3rd Prize
2018/2019 “Top 10 Bioinformatics Breakthroughs in China” by GPB
2019 “Top 10 Bioinformatics Algorithms & Tools in China”
PDCAT 2005 / ICIBM 2019 Best Article
Associate Editor of IEEE/ACM Trans. Comput. Biol. Bioinform
National Undergraduate Challenge Cup Extinguished Advisor

Leadership

Hainian ZENG, CEO
10+yrs Investment, BD, Fundraising, RA
10+yrs in healthcare industry regarding management, BD, investment and fundraising
Portfolio: Tmunity, NextCure, HUA Medicine, KBP, XGENE, Rani, Prenetics and etc.
Adjunct Reviewer/Inspector at Shanghai FDA/CDE, Certified Pharmacist
Yanhong ZHANG, BD VP
30+yrs BD, CRO Operation, Management
20+yrs C-level position in big pharma
Ten years CRO operation & management, strategic planning, BD and marketing, develop & maintain KA
Familiar with the Chinese healthcare industry, well-connected
Haiquan FANG, DD VP
20+yrs Drug Discovery, Medicinal Chemistry
20+yrs experience working in multinational pharmaceutical companies
Led and participated in several drug discovery programs with two NDA approvals, several at later stages and dozens INDs
Led drug discovery team in Chinese biotech companies

Silexon’s AI team ranks among Top Tier in the world

Illuminating the Druggable Genome (IDG)-DREAM Challenge & CTD2 Pancancer Drug Activity DREAM Challenge

#1 in High-impact Peer-reviewed Publications and Modeling Contests Leading Company in AI + Drug Discovery and Biopharma Research

20+
models
70+
papers
12
patents

Value Proposition

Proprietary AI Models can address biopharma R&D issues

Platform

We take pride in our AI4D™ platform and comprehensive AI models
AI4D™
AI
for Drug Discovery, Design & Development
-SAFE Selective, Accurate, Fast, Efficient
-Applicable in FIC, fast follow-on, me-too-me-better & repurposing for small molecules
-Multiple AI models/modules aimed at various “4D” issues
-Synergistic with emerging therapeutic modalities such as PROTAC, DEL and etc..

Discovery
-Proprietary virtual library to explore vast chemical space
-Customized virtual library with directed / focused molecules
-Virtual high-throughput screening
-Confirmatory screening with docking and MD

Development
-Knowledge graph
-Translational medicine
-Novel target ID & repurposing
-Real-world evidence and study

Design
-Automatic patent research
-Covalent structure assembly
-Interaction-based / SAR drug design
-Druglikeness & ADMET prediction

Models

Silexon has 4 various AI models for virtual high-throughput screening and Drug-Target interaction/affinity predictions.

Switch to know more models

Network based
Achieve substantial performance improvement over other state-of-the-art methods for drug–target interaction prediction
NeoDTI: a new nonlinear end-to-end learning model, which integrates diverse information from heterogeneous network data and automatically learns topology-preserving representations of drugs and targets to facilitate DTI prediction
Achieve substantial performance improvement over other state-of-the-art DTI prediction methods
Robust against a wide range of hyperparameters and ready to integrate more DTI information such as binding affinity
Sequence based
Sequence Model description: a novel general and scalable computational framework that combines effective feature embedding with powerful deep learning methods to accurately predict compound-protein interactions (CPIs) at a large scale, with superior predictive performance compared to baseline methods, capable to predict structure-free CPIs
Prediction Accuracy above 90%
Applicable in Virtual High-throughput Screening for FIC and difficult-to-drug targets, 3-6 hours for 1M molecule screening for drug discovery
Wet-lab validation can be reduced to the scale of 100 and fewer, with capabilities to identify allosteric hits/leads
Interaction based
Model description: a multi-objective neural network to predict both non-covalent interactions and binding affinity for a given compound-protein pair, with interpretability of the identified local features for compound-protein pairs, and provide useful insights about the molecular mechanisms of compound-protein interactions, help advance the drug discovery process
Applicable in first-in-class and fast follow drug discovery projects
Outperform other state-of-the-art methods in predicting compound-protein binding affinities
Next-gen Model
Model description: a next-generation multi-objective neural network to predict both non-covalent interactions and binding affinity for a given compound-protein pair, leveraging SBDD, FBDD, LBDD and interpretability of the identified local features for compound-protein pairs, and provide useful insights about the molecular mechanisms of compound-protein interactions, help advance the drug discovery process
Applicable in first-in-class and fast follow drug discovery projects
SILEXON aims at developing practical AIDD models

SILEXON’s solutions for challenges in AIDD

Model designs based on biological and medicinal chemical knowledge

Rich experience and systematic methodologies on biological tasks (including modeling sequences, structures, compounds, omics and network data and related problems)

Iteratively updated key models with superior performance and successful application cases;

Four generations of DTI models with high hit identification rates, for targets of all categories (first-in-class, difficult-to-drug targets, allosteric modulators, fast follow and repurposing)

Two developing DTI models with novel improvements: 1) an AI+CADD model learning from molecular dynamics (MD) to achieve both efficiency and accuracy (one million times speed-up); 2) an integrated model for SBDD+FBDD+LDBB+interaction

AI4Pat (automatic patent analysis) for extracting valuable data from patents: combined with high-throughput techniques and in vitro biochemical assays, the obtained data can be used to iteratively optimize the models (forming up a model-data-drug-model loop)

Don’t miss anything

News

In progress...