|
|
Dr.
Kwoh Chee Keong, BBM,
PBM, PBS, CRM PhD, DIC, MSc(ISE), Beng(EE), PGDIG, Senior Member, IEEE Senior Member IES Life Member ICAAS Member AMBIS College
of Computing and Data Science (CCDS) S3 B1C-91 Nanyang
Technological University (NTU) 50 Nanyang Avenue,
Singapore 639798 T: +65 6790 6057 F: +65 6792 6559 W: https://personal.ntu.edu.sg/asckkwoh/ |
I think the nicest, most sincere compliments that I have
received are those from my students and people I did not expect.
Notes from Students and Friends
HONORS AND AWARDS
National Day Awards
Bintang Bakti
Masyarakat, The Public Service Star (BBM)
National
Day Awards 2026
Conferred by the President of Singapore
Pingat Bakti Masyarakat, The Public Service Medal (PBM)
National Day Awards 2008
https://www.pmo.gov.sg/national-awards/recipients/
Conferred by the President of Singapore
Pingat Bakti Setia, The Long Service Medal
National Day Awards 2016
Ministry of Education
National Day Awards 2023
Ministry
of Culture, Community and Youth
Other Awards (NTU)
Best Faculty Mentor Award from
Temasek Foundation (TF) LEARN 2014
Best Faculty Mentor Award from
Temasek Foundation (TF) LEARN 2013
PUBLICATIONS
From Google Scholar
https://scholar.google.com.sg/citations?hl=en&user=jVn0wDMAAAAJ&view_op=list_works&sortby=pubdate
RESEARCH EXPERTISE
My main interests lie in our desire
to making sense of big heterogeneous data for real
application in engineering, life science, and medical.
Data Science
(Bioinformatics) Research
High throughput
biological measurements and experiments in life science and healthcare have resulted
in the explosion of data available from sequencing and micro-arrays,
ChIP-Microarrays (ChIP-chip). This has led to the interdisciplinary science
called Bioinformatics. Which use Data Science in solving biological and life
science problems.
Machine Learning and Statistical Inference
Machine Learning
and statistical modelling techniques that
can learn from data to enables the making of the decision and simply a classification, this has application in almost every area. There
are many approaches and each has its own merits. The following have been
heavily used in my group: support vector machines, decision tree learning,
artificial neural networks, Bayesian networks, genetic and mimetic algorithms. Example
include application in supertype-specific HLA Class I binding peptides, protein-ligand binding affinity.
Learning with Unlabeled Data
In the context
of machine learning, PU learning is a collection of semi-supervised techniques for
training binary classifiers on positive (P) and unlabeled (U) examples. To
improve performance, it is important to partition U into four sets, namely,
reliable negative set RN, likely positive set LP, likely negative set LN, and
weak negative set WN. Such an approach
has been used to identifying disease genes from the human genome is an important but challenging task in biomedical
research.
Meta and Ensemble learning
Meta-learning
is where automatic learning algorithms are applied to meta-data about machine learning experiments. The main goal is
to use meta-data to understand how automatic learning can become flexible in
solving different kinds of learning problems and enrich the knowledge
discovered. Coupled with ensemble methods that that integrates results of multiple predictive methods into one
system, these approach has found to be instrumental in
improving predictive performance.
Application of this approach has been widely used in big data such as
bioinformatics and medical informatics. An example includes multiple kernel
learning for heterogeneous data fusion and sparse learning in genome-wide association study (GWAS), and drug-target
interaction prediction.
Ontology for
Knowledge Representation
Ontology is a
formal representation of a set of concepts within a domain and the
relationships between those concepts. There have been major initiatives in
medical and bioinformatics to standardize the representation of terminology and
relationship across diseases, species and databases.
The controlled vocabulary of terms and description of product characteristics
are the main outcome to improve annotation for representation of meta-concept that greatly enhances analysis. Specifically, it has been extensively applied in complex mining,
inferring gene-disease-phenotype association.
Computational
Structural Biology
This is where
Computational Science and Engineering finds its
applications in understanding the molecular
structure of biological macromolecules. Dealing with computational models and
simulations, the field makes use of high-performance computing to solve complex
and expensive problems in biology. The result
includes the acceleration of docking and
its application in the understanding of
influenza and other viruses.
SOFTWARE
CovalentDock Cloud: a web server
for automated covalent docking
Covalent binding is an important mechanism for many drugs to
gain its function. We developed a computational algorithm to model this
chemical event and extended it to a web server, the CovalentDock
Cloud, to make it accessible directly online without any local installation and
configuration. It provides a simple yet user-friendly web interface to perform
covalent docking experiments and analysis online.
Software for Accelerating Autodock
Vina
Quickvina: This project
aims at accelerating Autodock Vina, a program for
protein-ligand docking. The main idea is to skip some of the local searches
which are not promising in finding a
better solution.
Quick
Vina 2 is a fast and accurate molecular docking tool, attained at
accurately accelerating AutoDock Vina. It was tested
against 195 protein-ligand complexes that
compose the core set of the 2014 release of the PDBbind
using default exhaustiveness level of 8, QVina
2 successfully attained up to 20.49-fold acceleration over Vina.
Software for Learning with Unlabeled Data
1.
PUDI (2013) - a Positive-Unlabeled (PU) learning based method
aiming to address the problem of disease gene identification
Software – for complexes
1.
CACHET- Discovery of Protein Complexes with
Core-Attachment Structures from TAP Data
2.
COACH- COre-AttaCHment
based Complex Mining
Software – for Computational Structural Biology
1.
CovalentDock Cloud (2013) - This web server allows the researchers and scientists to perform protein-ligand covalent
docking.
2. CovalentDock: Automated covalent docking with parameterized
covalent linkage energy estimation and molecular geometry constrains
3.
QuickVina: Accelerating AutoDock Vina Using Gradient-based
Heuristics for Global Optimization
MY
GRANTS
· Host-pathogen protein-protein
interaction approaches for predicting virulence
· The discovery of neutralizing
antibodies for potential novel coronavirus through machine learning approaches
· Explainable AI for Multimodal
Predictive Maintenance of Jet Engines with Smart HCI
· Hybrid Finite Element Method And Mixedlevel Coarse GrainingMolecular Dynamics Simulation
· Computational Virulence Model With Functional Information For
Influenza Viruses
· Structural analysis and
characterization of protein complexes
· Towards direct and rapid mapping of
RNA modifications with nanopore sequencing
· Untangling cancer re-wiring:
Pan-Cancer mapping of transcription factor driven dysregulatory
hotspots using AlphaFold2 and integrative machine learning
· Investigating the regulation of 3D
genome organization using machine learning
· Predict the solubility of proteins
using machine learning
· Hodge Laplacian based deep learning models
for drug design
· Challenge-Learn: Developing and
Assessing an Andragogical Programme and System based on Co-Skilling to Enhance
Employability and Learning
· Artificial intelligence for the
prediction of alternative splicing from epigenomics and transcriptomics data in
cancer
· Computational Systems Biology of
Synthetic Lethality Towards New Cancer Medicine
· Untangling cancer re-wiring:
Pan-Cancer mapping of transcription factor driven dysregulatory
hotspots using AlphaFold2 and integrative machine learning
· Predict the solubility of
proteins using machine learning
· Hodge Laplacian based deep
learning models for drug design
· Challenge-Learn: Developing
and Assessing an Andragogical Programme and System based on Co-Skilling to
Enhance Employability and Learning
· Host-pathogen
protein-protein interaction approaches for predicting virulence
· The discovery of
neutralizing antibodies for potential novel coronavirus through machine
learning approaches
· Artificial intelligence for
the prediction of alternative splicing from epigenomics and transcriptomics
data in cancer
· Structural analysis and
characterization of protein complexes
· AI Enhanced Creativity In Education
· Hybrid Finite Element Method
And Mixedlevel Coarse GrainingMolecular Dynamics Simulation
· Computational Virulence
Model With Functional Information For
Influenza Viruses
· Computational Systems
Biology of Synthetic Lethality towards New Cancer Medicine
· CloudDock: Molecular Docking Platform
on Cloud
· Methodological Investigation
for Automatic Detection of Primary Angle Closure Condition (PAC) and PAC
induced Glaucoma
· Bioinformatics Algorithms
for Detecting Genetic and Epigenetic Determinants of Meiotic Recombination
Hotspots from Genomic Data
· Core-Attachment based Mining
Technique: to detect Protein Complexes and Protein-Small Molecule Interactions
· Core-Attachment based Mining for Protein Complexes & Small-molecule
Interactions
· Improved Design via Evoltionary Algorithms
· The
Protein Binding Hot Spots Are Water Free?
· Neural Systems Modeling with
functional MRI
· Function MR Time-Series
Analysis
· Augmented reality for
prosthesis cup placement
· Cardiovascular &
respiratory systems' signal simulation, processing and analysis for ICU, or and telemedicine applications.Computational
Virulence Model with Functional Information for Influenza Viruses
· Protein
binding hotspots are water-free?
· Analysis of Past DRG data
for the study of LOS for better utilization of Hospital Resources
· Data Warehousing and Data Mining
Analysis of Staphylococcus Aureus
· A novel approach for inter-
to intra- network analysis of genetic
diseases using high-throughput data
· Neural Systems modelling with functional MRI
· SCE incubator proposal for
“Evolutionary and Complex Systems Lab”
· The Application of ultrasound-based augmented reality with the
directional vacuum-assisted breast biopsy device in the treatment of breast
cancer
· Distributed Diagnosis and
Home Healthcare (D2H2)
· Development of a robotic semi-automated
remote handling system for radioiodine dispensing
· Functional MR Time-Series
Analysis
· Augmented Reality for
Prosthesis Cup Placement
· Robotic Skull Based Surgery
· Cardiovascular and
Respiratory Systems' Signal Simulation, Processing and Analysis for ICU, OR and
Telemedicine Applications.
· Strategic research:
Interventive augmented reality for medical applications.
· Surgeon Assistant Robot for
a Selected urological disorder.
MY
GRADUATE STUDENTS
· Tan Lai Heng
· Emadeldeen Ahmed Ibrahim Ahmed Eldele
· Zhang Yu
· Mohamed Ragab Mohamed Adam
· Lin Zhuoyi
· Hou Yubo
· Tjio Ci'en
Gabriel
· Li Xinya
· Yin Rui
· Zhou Xinrui
· Ata Kircali
Sezin
· Amr Ali Mokhtar Alhossary
· Aly Mohamed Alaaeldin Aly Ezzat
· Pradhan Mohan Rajan
· Pan Hong – DNA methylation
biomarkers of personal disease risk (PhD, 2012)
· Luay Aswad - A molecular basis of
the 5-gene breast tumour aggressiveness
grading signature (AGS) and its network – PhD, (2012 -)
· Han Xu - Constructing the Semantic
Web for Biomedical Literature (PhD, 2011 - )
· Ouyang Xuchang - Automated and
Accelerated Covalent Docking and Covalent Virtual Screening (PhD, 2010–)
· Thidathip Wongsurawat
- Computational Analysis and Prediction of Specific Genomic Regions Forming
R-loop Structure and Chromosomal Variations Associated with Cancer - (PhD, 2015)
· Zhang Zhou - Knowledge Discovery In Post Genome-Wide Association Study For
Glaucoma (PhD, 2015)
· Su Tran To Chinh - Improving the
Discrimination of Near-Native Complexes for Protein Rigid Docking by Implementing
Interfacial Water into Protein Interfaces (PhD, 2015)
· Yang Peng - Computational Approaches
for Disease Gene Identification (PhD, 2014)
· Wu Min - Mining Protein Complexes From Protein Interaction Data (PhD, 2012)
· Zhang Tianyou - Contact Network
Based Framework For Infectious Disease Interventions
(PhD, 2015)
· Stephanus Daniel Handoko -
Constrained-Oriented Refinement-Efficacious Memetic Algorithms for Efficient
Optimization of Computationally-Expensive Problems
(PhD, 2014)
· Adrianto Wirawan - Whole-Genome Discovery Of Transcriptional Regulator Binding Sites (PhD, 2011)
· Zhang Guanglan- Computational
Epitope-Driven Vaccine Design (PhD, 2008)
· Zheng Yun- Design Of
Gene Expression Networks From Microarray Data (PhD,
2006)
· Zhao Ying- Efficient Model And Feature Selection For SVM In
Biomedical Data Analysis (M Eng, -2004)
· Zhao Jianhui-
Human Animation from Motion Recognition, Analysis and Optimisation ( PhD, 2003)
· Chen Yintao
- Image Processing For Ultrasound Guidance System In Breast Lump Operation (M Eng, 2002)
· Wang Yan - Image-Based Indexing And Retrieval Of Trademark Logos,
(M Eng, 2001)
· Veena Mohan Bhajammanavar
- Image Processing Of The
Digital Mammogram For Segmentation And
Characterization Of Microcalcifications, (M Eng, 2000)
· Misra Sabita - Time Series Analysis Of ECG For Detection Of Premature
Ventricular Contraction (M Eng, 2000)
· Zou Qingsong
- Object-Based Volume Visualisation For Medical Imaging (PhD, 2001)
Planned and lectured subjects in
- CZ4032 Data Analytics and
Mining (2014,15): Data Mining is an analytic process designed to
explore big data in search of consistent patterns and/or systematic
relationships between variables, and then to validate the findings by
applying the algorithms to new data.
- CE7411 Bioinformatics
(2015): This course covers basic
bioinformatics concepts, databases, tools and applications. Introduction:
cell biology's central dogma, biological technologies for collecting and
storing genomic sequence data; databases that store these data and
strategies to extract information from them; Pairwise sequence alignment
for assessment of similarity to infer homology; Fundamental of Scoring
matrices to understand the assigned scores when performing alignment; The
popular heuristic search tool - Basic Local Alignment Search Tool (BLAST)
and advanced database searching; Multiple sequence alignment and
phylogenetic trees to complete the coverage from genomic sequences.
Functional genomics with the introduction to gene expression. Processes
for microarray data analysis; Feature selection and classification for
microarray data analysis. Protein families & proteomics; Protein
structure and structural genomics; and Molecular evolution and phylogeny.
- BI6123 Methods and Tools
of Proteomics (2007): Proteomics study and
identify protein structure, interactions of protein/protein and
protein/DNA and biology of organisms. We will further introduce the newly
developed technology for the quantitative analysis of protein expression
and function on a genome-wide scale.
- BI1602 and SC448
Introductory Bioinformatics (2005,06): Basic
bioinformatics concepts. Databases, tools and applications.
- BI1603 Computational Biology (2006) Introduce the
applications of the techniques of computer science, applied mathematics,
and statistics to address problems inspired by biology. Major
computational techniques used in biology include
Bayes, HMM, MI etc.
- BG3011 Biocomputing
(2005, 06): Introduction
the new course of biocomputing for students in SCBE, the subject is first
offered in July 2005; It covers Concepts; Bioinformatics databases;
Sequence alignment; Phylogeny and protein structure prediction.
- BI6104 Biostatistics: First offered in July 2003, this course equipped the
students in MSc with Knowledge of statistics, experimental design and
statistical learning.
- Curriculum for MSc in
Bioinformatics: From August 2001 to
June 2002, I worked with Vice-Dean (Academic) SCE, Head, Natural Science
of NIE, Vice-Dean (Academic) of SBS and Professor from MPE and EEE to
structure the new MSc in Bioinformatics.
- SC104 Mathematics I Fundamental of mathematics for Engineering
include statistics and calculus
- CE307 Computer
Peripherals: In 1996-2000, re-design the course to
include start-of-the-art techniques such as PRML, USB and Bluetooth.
- M495 & M6524 Medical
Assist Surgery (2000-2002): Co-planed and lectured the final year and MSc
elective for Biomedical Engineering.
- Digital Signal
Processing (1992): Planed and
lectured the final year elective for the computer engineering.
MY PHD THESIS
GRADUATE ADVISORS: Prof Duncan Fyfe Gillies - Professor
of Biomedical Data Analysis, Department of Computing, Imperial College London
My
PhD thesis Probabilistic
Reasoning From Correlated Objective Data,
University of London, Imperial College