Embedded Files
Today's society is facing difficulties such as chronic and aging diseases associated with an aging population, as well as overuse of resources. Only through cross-border communication and cooperation between experts in cross-border fields with sharing technology and resources, the medical problems can be actually solved. Thus, APBA brings together various core technologies including the follow.
BT&D² Medical & Pharmaceutical R&D System
BT&D² Medical & Pharmaceutical R&D System
Taiwan BT&D² Team
PI：Prof. Hsien-Tai Chiu, National Cheng Kung University
For unmet needs & markets, BT&D2 System integrates AI, Big Data, and Combined Medicine to provide products & services of Chinese and Western medicine. BT&D2 System integrates 4 core databases (Disease Target, Drug, Drug Modification Enzyme, Evidence-Based Medicine), utilizing AI operation system to develop new drugs. It not only matches disease targets with drugs for paired analysis and prediction but also identifies medicine for disease.
"AI, Big Data, and Combined Medicine" from BT&D2 Medical & Pharmaceutical R&D System
Drug Modification System
Drug Modification System
Taiwan BT&D² Team
PI：Prof. Hsien-Tai Chiu, National Cheng Kung University
Drug Modification System (DMS) is distinguished by its environmentally friendly, combined biosynthesis process and the Taiwan BT&D² Team's unique AI-driven technology, which together enable the design of highly effective and feasible drugs. These drugs, along with their various derivatives, are synthesized for applications in disease treatment and agriculture. The process utilizes natural biosynthetic enzymes to generate the functional groups needed for drug structure modification, and modifying enzymes to fine-tune these groups. Through the Taiwan BT&D² Team’s innovative in-situ one-pot (iSOP) process, the enzyme tandem platform integrates a modular approach, mixing and matching different enzymes to create multiple synthesis pathways. This allows for the efficient production of a wide range of drugs, maximizing economic benefits in a single step.
"AI, Big Data, and Combined Medicine" from BT&D2 Medical & Pharmaceutical R&D System
Traditional Clinical Medical & Pharmacopoeia Intelligence Analysis System, TCMPIAS
Traditional Clinical Medical & Pharmacopoeia Intelligence Analysis System, TCMPIAS
Taiwan BT&D² Team
PI：Prof. Hsien-Tai Chiu, National Cheng Kung University
TCMPIAS is established by similar principles adapted by BT&D2 AI-based drug-targeting analysis system, where extensive data from Western empirical science, including basic and clinical medicine, are integrated and analyzed. In particular, TCMPIA is developed in an innovative approach by integrating and AI-analyzing the meridian system and internal organ physiology of TCM with pharmacological insights from Compendium of Materia Medica of Chinese Pharmacopoeia. TCMPIAS harmonizes Western and Eastern medical sciences in a systematic manner and allows to predict effective formulas for treating diseases.
Live-Virus Neutralization Assay Platform
Live-Virus Neutralization Assay Platform
Live-Virus Neutralization Assay Platform enables the study of significant viral pathogens, including Influenza A (H5N1), HIV, and SARS-CoV-2. The platform is highly versatile and can be applied to various research areas, such as tracking the long-term dynamics of antibodies against SARS-CoV-2 in recovering COVID-19 patients.
SARS-CoV-2 Human
SARS-CoV-2 Human
Primary Cell Infection Platform
Primary Cell Infection Platform
Motivated by the need for crucial therapeutics during the COVID-19 pandemic, AT Virology Lab aimed to establish organ-specific models including mesenchymal stem cells and cardiomyocytes for the study of viral infections and develop them into novel platforms for drug discovery.
High-Content Screening Platform
High-Content Screening Platform
for Antiviral Candidates
for Antiviral Candidates
Using high-content screening for antiviral candidates by using fluorescence-based SARS-CoV-2 nucleoprotein detection, we demonstrated that the Thai medicinal herb, were promising candidates for the novel treatment against COVID-19.
We were also able to evaluate the anti-SARS-CoV-2 activity of Andrographis paniculata extract and its major component, andrographolide.
Study Design of Dose-Dependent Anti-SARS-CoV-2 Effects.
CURATE.AI
CURATE.AI
CURATE.AI optimizes personalized medicine with srtificial intelligence. Through calibration of patient data from an initial physician-guided six-dosing block period, CURATE.AI generates a personalized profile using a second-order quadratic correlation between drug doses (input 1, input 2), and a phenotypic patient response (output). The generated second-order equation contains patient-specific coefficients (C1, C2, C3, C4, C5, C6), which CURATE.AI then analyzes to predict dose combinations that guide patient responses toward a preferable outcome. CURATE.AI recalibrates the generated personalized profile using the latest available patient data to accommodate changes in patient state.
CURATE.AI-Guided Dosing Workflow for a Two-Drug Optimization.
IDentif.AI Online
This resource is based on a recently completed study that harnessed the IDentif.AI platform to experimentally pinpoint a broad spectrum of potential combinations against SARS-CoV-2. IDentif.AI does not use in silico modelling or synergy predictions. Instead, it pairs prospective experimental validation with an optimisation process to provide a list of regimens, which can be explored through this interactive resource, for further consideration. This database will be updated as additional candidate therapies are assessed.
Mapping Analogous Nuclei Onto
Mapping Analogous Nuclei Onto
Residue And Affinity: MANORAA
Residue And Affinity: MANORAA
The key inventions are around the webservice Manoraa.org systems to assist drug discovery by linking ligand to target proteins, baseline expression, SNPs, pathways, tissues, and organs. By linking these information in this Manoraa ligand design hub, researchers can perform in silico target discovery and ligand design before planing their wet lab experiments. I have also developed algorithms to analyze conserved features obtained from a set of homologous protein crystal structures which can guide inhibitor design. It enables us to compare the shape of the pocket, to observe position specific interactions, and to display chemical interactions in the pocket of the proteins.
Deep Neural Network-Assisted Drug Recommendation Systems, dNNDR
Deep Neural Network-Assisted Drug Recommendation Systems, dNNDR
A machine learning-based multiclass classification workflow that segregates interactions between active, inactive, and intermediate drug–target pairs. Drug molecules, protein sequences, and molecular descriptors were transformed into machine-interpretable embeddings to extract critical features from standard datasets. Tools such as CHEMBL web resource, iFeature, and an in-house developed deep neural network-assisted drug recommendation (dNNDR)-featx were employed for data retrieval and processing. The models were trained with large-scale DTI datasets, which reported an improvement in performance over baseline methods.
Electrophysiological Core &
Electrophysiological Core &
In vivo Manipulation Core
In vivo Manipulation Core
Laboratory Of Neurodevelopment And Neuroepigenetics
PI：Prof. Hsien-Sung Huang, National Taiwan University
Laboratory Of Neurodevelopment And Neuroepigenetics perform the whole-cell patch clamp technique to record neuronal intrinsic excitability (Fig. a) and synaptic transmission (Fig. b) from mouse brain slices. In addition, we apply the extracellular field potential recordings to measure synaptic plasticity such as long-term potentiation (LTP) (Fig. c) from mouse brain slices. These techniques can help researchers address questions in a circuit level.
Besides, the Laboratory use adeno-associated virus (AAV) as a vehicle to carry our target DNA constructs, deliver AAV to specific mouse brain regions with stereotactic surgery, manipulate gene expression with rescue or CRISPR approach in vivo, and exploit neuronal activity with optogenetic (Fig. a & b) or chemogenetic (Fig. c & d) techniques in vivo. These techniques can help researchers address questions in a causal manner and under in vivo context.
Core for measurement and manipulation of brain circuit dynamics in free-moving mice
Laboratory Of Neurodevelopment And Neuroepigenetics
PI：Prof. Hsien-Sung Huang, National Taiwan University
Laboratory Of Neurodevelopment And Neuroepigenetics use the nVoke system (Inscopix) to measure and manipulate brain circuit dynamics in free-moving mice. Specifically, we perform in vivo cellular-resolution calcium imaging with simultaneous or sequential opotogenetic manipulation in free-moving mice brain. This technique can help researchers to address the causal link between neural circuit and related behaviors.
Mouse Behavioral
Core & Epigenetic core
Laboratory Of Neurodevelopment And Neuroepigenetics
PI：Prof. Hsien-Sung Huang, National Taiwan University
Laboratory Of Neurodevelopment And Neuroepigenetics evaluate behavioral outputs from mouse model of human brain disorders within five categories of learning and memory, social interaction, sensorimotor activity, anxiety, and depression. These techniques can help researchers address questions in a comprehensive manner and provide them the highest level of functional readouts of brains.
Also, the Laboratory establish a hybrid mating system with B6 and CAST mouse strains to determine levels of gene expression (Fig. a) and DNA methylation (Fig. b) in a parent-of-origin-specific manner.
Binding INteraction Determination, BIND
Binding INteraction Determination, BIND
The BIND model for Structure-Free Virtual Screening
BIND is able to perform forward screening, reverse screening, and drug-target affinity prediction, all without structural input. BIND is trained on the BindingDB dataset only consisting of protein sequences and experimentally determined DTA values. By attaching BIND to a pre-trained ESM-2 protein language model and feeding in a protein sequence and SMILES molecular representation, which are then converted into graphs, the model can effectively discriminate between active and decoy ligands. This allows BIND to be used in forward and reverse screening, while predicting DTA values.
Page updated
Google Sites
Report abuse