Project Collab Directory

Computing with artificial and biological neural networks

My main research interests revolve entirely around Neural Computation. Together with several PhD students and collaborators, I am trying to unravel what computations are being carried out by different neural systems, and how and why they are being implemented. Therefore, I am interested in both biological and artificial aspects of neural computation.

On the biological side, my work falls mostly within the field of Computational Neuroscience concerned with how different biological neural systems solve different types of computational problems. In terms of problem domains, I have mostly been involved in the research of visual functions and therefore Computer Vision and Image Processing are central to my investigations. I am currently focusing on the simulation-modelling of the biological retina, through which I hope to:
1) answer several questions pertaining to the structure and function of different retinae,
2) propose new image processing techniques for low-level functions such as illumination normalization and colour correction, and
3) propose new designs for retinal prostheses.

So far, our models of the outer retina have revealed that the relatively simple micro-circuits established between photoreceptors, horizontal cells and bipolar cells are capable of a significant number of distinct image processing functions such as denoising, contrast and saturation enhancement, edge detection and colour normalization (see Figure 1 for an example of local illumination normalization and denoising). We are currently in the process of extending our models of the retina, by adding several amacrine and ganglion cell types, with the aim of understanding retinal colour coding and processing. This understanding should guide us towards achieving the goal of enhancing retinal prostheses for allowing users to experience coherent colour perception.

On the artificial or applied side we are developing new types of hybrid artificial neural networks with a particular emphasis on functional diversity, i.e., Neural Diversity Machines (NDM). The notion of diversity is borrowed from biological neural systems where we observe a significant diversity of neurons in terms of morphology, connectivity, electrophysiology, and other properties. The case of the retina, where on average (across species) there are approximately 55 different types of neuron, is one simple illustration of this point. Initial experiments have shown some promising results and our current priority is to improve the speed and reliability of optimization methods for NDM.

Although NDMs are intended to be general and have already been tested on several unrelated benchmarks from the UCI Machine Learning Repository, one of the target domains which ties NDMs with our more biologically oriented research pertains to the segmentation of histological images of the retina for automated or semi-automated reconstruction of retinal circuits (i.e. connectomics research). The motivation behind this application is to bridge one gap in the chain from biological specimens through data acquisition up to computational modelling. To completely bridge the gaps in this chain means that one day we will be able to scan a neural tissue (e.g., retina) and automatically generate computational insights. As collaboration is essential (e.g., biologists and computer scientists), I look forward to receiving any potential queries from interested parties.

Figure 1: Examples of image processing by a neural model of the outer retina.

Scientific Malaysian Profile:
http://www.scientificmalaysian.com/scimy/members/tomasmaul/profile/

Links:
[1] Research page. http://baggins.nottingham.edu.my/~kcztm/Research.html
[2] Interdisciplinary Computing and Complex Systems research group. http://icos.cs.nott.ac.uk/
[3] Cognitive and Sensory Systems research group. http://www.nottingham.edu.my/Psychology/Research/CognitiveandSensorySystemsGroup.aspx

Earthquake geology of Southeast Asia

In many parts of the world, active faults are too poorly understood to allow even rudimentary quantification of their role in regional neotectonic deformation or to make realistic estimates of seismic hazards, let alone seismic risk assessments.Such was the case before the infamous 1999 Chi-Chi earthquake in Taiwan. That rather late awakening in Taiwan to the value of using tectonic landforms in the characterization of active faults and folds led to the systematic application of geomorphic evidence to characterize the active tectonics throughout the island and offshore, which was carried out by J. Bruce H. Shyu (National Taiwan University, Taipei, Taiwan) and his group. Although the active fault that produced the earthquake had long been known from the interpretation of subsurface (stratum below the earth’s surface) data and outcrops (group of rocks that stick out of the ground), it had not been considered active. Surface fault ruptures and associated folding during the earthquake clearly revealed the fault’s location, danger and clear geomorphic evidence for recent, prior activity. A similar geomorphological investigation of the neotectonic elements of Myanmar is largely completed, as part of the Ph.D. work of Caltech graduate student Wang Yu.

The situation in western Indonesia after the 2004 Indian Ocean earthquake and tsunami illustrates well the state of SEAsian neotectonics. Soon after the devastating earthquake and tsunami of 2004, when theUnited States Geological Survey(USGS)was producing a Probabilistic Seismic Hazard Analysis (PSHA) of the western half of Indonesia,survey personnel found that there was little in the way of reliable active-fault mapping. We, at the Earth Observatory of Singapore,were asked to rapidly compile a map of the active faults of Sumatra, Java, Borneo and Bali to serve as a key element in their evaluation of probabilistic earthquake hazard. We spent a couple of months evaluating Shuttle Radar Topography Mission (SRTM) digital imagery and presented them with a very crude map of active structures, which they used in their analysis.  

The active tectonic elements of many other regions within SE Asia are also not well understood. These include eastern Indonesia, Malaysia, Laos, Vietnam, Cambodia and many offshore tracts. Likewise, rates of fault slip and past seismic history are for the most part poorly known. This regional state-of-knowledge contrasts markedly with better-studied places such as Japan and California. More than two decades ago, earlier thorough mapping of active structures enabled creation of the first probabilistic seismic hazard maps of those places. The fact that nearly all of the dozen or so subsequent destructive earthquakes in California have occurred within the high-probability areas shown by the California maps attests to the reliability of these early PSHA maps. More recent PSHA maps covering the entirety of the US are the basis for today’s seismic zonation of the country. These maps, too, are based upon knowledge of the location of active faults and varying levels of understanding of their slip rates.

The Earth Observatory of Singapore was established in 2009 to fill the above-mentioned gaps in research and the aim is “to conduct fundamental research on earthquakes, volcanic eruptions, tsunami and climate change in and around SE Asia, toward safer and more sustainable societies”.It has successfully conducted research in most of the SE Asia. However, to the best of our knowledge, there has not been involvement of any scientist from Malaysia to conduct the earthquake research. We are aware that the need to undertake a thorough research study on earthquakes is more in Malaysia than in Singapore, potentially because of the tectonic settings and the dimensions of the country . This has not been done in any part of the country and therefore, it is extremely important to map all the active tectonic features in Malaysia.By observing the satellite images, many places in Malaysia with active faults should be mapped in greater detail, so that we can understand the potential of earthquake risk in the country and surrounding areas in the future. Also, the potential earthquake risk from the neighbouring regions (e.g. Indonesia) should be assessed. There are various seismic sources around this country (Figure 1, upper panel), which are partially mapped and based on these structures,the USGS has produced a generalised earthquake hazard map of the region (Figure 1, lower panel). Though it should be noted that this map is crude and therefore, it is compulsory to conduct a thorough research on risk of earthquakes in order to generate a more realistic earthquake hazard map of the region. Also, the paleoseismic (old earthquakes) research is required to understand the past earthquakes and their sources.


Figure 1: The tectonic setting (upper panel) and generalized seismic hazard (lower panel) of Malaysia and its surrounding areas

As a conclusion, the earth sciences education is extremely important to minimize the risks associated with natural hazards. Unfortunately,Malaysian higher education institutions do not offer a comprehensive geosciences degree program. To fill these obvious gaps, both in education and research, will require serious collaborations with earth scientists around the globe and therefore, we would be happy to work with any of the experts in Malaysia to spread the earth sciences awareness in this country.

Scientific Malaysian Profile:
http://www.scientificmalaysian.com/scimy/members/afrozshah/profile/  

Link:
TEDxNTU talk by Dr Afroz Ahmad Shah – Is earthquake prediction a possibility? 

Nutraceutical powder and beverages

Furley Group of Companies started as a humble food ingredient trader comprising of two individuals. As the customer base grew, so did the demands for expertise in formulations for nutraceutical powders and beverages. To cater to this demand, Furley Bioextracts was formed in 2008. Furley Bioextracts has strong roots in the nutraceutical beverages and our fortes are in fruit juice concentrates and powder blending. With assistance from the Malaysian government, Biotech Corp awarded Furley Bioextracts the BioNexus status as recognition of Furley Bioextracts’ involvement in biotechnology. This helped Furley establish itself as a Biotechnological industry player.

Not only does Furley do formulations, we also do manufacturing and establishment of quality systems beyond that of ISO 22000 for selected food products. This is to ensure consistency and safety of food products manufactured by Furley. Furley has two manufacturing sites which segregate manufacturing based on ingredient and customer requirements, be it environmental control, to specific laboratory or testing for quality control. Our manufacturing certifications include BioNexus Status, GMP, HACCP JAS-ANZ and Traditional cGMP, making us export ready. We are also working towards ISO 22000 and FSSC 22000.

Our research and development team comprises of 6 members with a majority of the members holding postgraduate degrees. Marketing and Sales comprises of 5 members and manufacturing comprises of 25 – 50 staff. This makes Furley Bioextracts a relatively tight knit team. Having decades of experience onboard, we are the preferred premium nutraceutical house working with food industry giants, from Swiss multinationals to small medium enterprises. We constantly strive to produce innovative premium products which have set us apart from our competitors.

In short, a successful formulation is not a matter of combining ingredients from a base formulation. A variety of factors has to be considered in order to achieve commercial viability. Some important issues to be considered include supply chain management, ingredient reactions, stability, sensory, legislation and product shelf life. We have collaborations with Forestry Research Institute Malaysia (FRIM) since late 2006 in the commercialization of their standardized herbal extracts. This collaboration has been given numerous awards (Geneva & ITEX), with its winning products ranging from soft serve ice-cream, beverages to cookies. Our latest milestone is the cultivation, processing and supply of mangosteen and local Malaysian herbal derivatives and extracts.

We are seeking to collaborate with institutions for the following purposes:
1) To conduct research they feel is feasible in the context of beverages or powders. We will provide assistance with commercialization of such products.
2) To standardize methods of supplying mangosteen and selected Malaysian herbs.
3) To conduct quality control processes for mangosteen and selected Malaysian herbs.
4) To conduct in-house quality control systems for manufacturing. We wish to receive product knowledge, marketable product, market exposure and recognition from the collaboration.

In return, the institute/scientist will receive:
i) Acknowledgement where deserved.
ii) Publications.
iii) Commercial prototypes.
iv) An advantageous edge in competitions for Awards e.g. Geneva or ITEX
v) An export ready product
vi) Satisfaction of consuming their research literally.

Depending on the feasibility of the end collaboration it may conclude with just a prototype (pure learning exercise) or a full commercialization process. This varies according to how the respective institution operates according to their terms and conditions. We are flexible and look forward to input from potential collaborators.

Scientific Malaysian Profile:
http://www.scientificmalaysian.com/scimy/members/kelvinsoo/profile/  

Neuroscience research in NBGG, Universiti Putra Malaysia

Neuroscience is one of the escalating and most fascinating endeavours of biology research today. Neuroscience research has advanced knowledge on how a brain functions, how a neurone behaves and how damaged neuronal networks may lead to various neurological disorders. In Universiti Putra Malaysia (UPM), neuroscience research is now one of the priority niche areas. Besides, steps have been taken to gather both the scientists and clinicians who venture in the field by pooling resources and practicing knowledge sharing. Our research group, known as NeuroBiology and Genetics Group (NBGG), is interested in unraveling the role of genetic factors and molecular networks that regulate the development and function of the mammalian brain. Our group place a great emphasis in three main areas of research; (1) neurological disorders, (2) non-coding RNA roles in brain development and function and (3) technology transfer and development on gene delivery platform. 

NBGG is involved in deciphering the genetic landscape leading to disrupted molecular pathways and processes responsible for Down syndrome pathology (trisomy 21) and associated disorders (defective neurogenesis and intellectual disability). With limited access to Down syndrome patient tissues and the lack of comprehensive investigations at the molecular level, we have employed Ts1Cje, a mouse model for Down syndrome to facilitate genetic dissection of the learning, behavioural and neurological abnormalities in Down syndrome. In collaboration with Professor Hamish Scott from the University of Adelaide, South Australia, we profile gene expression pattern at various regions of the Down syndrome brain at different stages of development. Spatiotemporal comparisons of gene expression profiles between the normal and Down syndrome brains provide a great genetic overview that may provide clues on what has gone wrong in the Down syndrome brain. In addition to the brain development, our interest also extends to identifying the molecular mechanism responsible for hypotonia (decrease in muscle tone), a cardinal feature, in Down syndrome. Our group plans to generate a comprehensive catalogue of molecular and cellular properties that affect locomotor functions and vesicle recycling mechanism at the neuromuscular junction of the Ts1cje mouse model. The ultimate aim of our research is to determine the effect of the additional gene dosage in the trisomic model. In the long term, NBGG aims to develop molecular therapies that may improve the quality of life among Down syndrome patients.

LEFT Cultured hippocampal neurones (in green) from the mouse.
RIGHT A cross section of a developing mouse brain.

A different branch of NBGG research is to understand the role of non-coding RNAs in regulating the development of the mammalian brain. Our group has a special interest in characterising the molecular role of a few novel microRNAs (miRNAs), that are found to be expressed throughout embryonic development especially in the brain. miRNAs are short RNA sequences with 18-24nt in length. miRNAs target mainly at the 3’ UTR of mRNA to either repress translation processes or promote mRNA degradation. In both cases, miRNAs will influence the level of protein synthesis. When the phenomenon happens at a global scale, changes of the amount of proteins synthesised may affect the phenotypic characteristics of the cell. In the context of a developing brain, miRNAs may play a crucial role in regulating neurogenesis, neuronal differentiation and function. To dissect the molecular role of these novel miRNAs, NBGG plans to use various techniques such as in situ hybridisation, stemloop-RT-qPCR, overexpression studies, Western blotting and luciferase assay analysis in various models such as mouse embryonic stem cells, differentiated neurones, primary neuronal cultures and brain sections.

To complement our efforts in elucidating the role of various candidate genes and miRNAs in brain development or function, our group is in the midst of establishing a “super electroporator” platform for the delivery of charged particles into specific regions of the mouse brain. Charged particles such as DNA constructs with Green Fluorescent Protein (GFP) as traceable gene reporter system have been established in our laboratory and are ready to be electroporated into the developing mouse brain in utero. The platform will facilitate us to trace the formation of defective migratory routes of neurones in the Down syndrome brain in the presence or absence of selected candidate trisomic genes. The platform also enables us to study the function of novel miRNAs in a spatiotemporal manner throughout the development and maturation of the brain.

Our group also works with clinicians to screen for aberrations of selected genes in Parkinson’s disease and infantile spastic patients. In addition, we look forward to any queries regarding our research projects and welcome any potential collaborators to work on the following extended research topics:

a) Advancement of the ‘super electroporator’ platform and its applications in in utero, in vitro as well as in vivo delivery of charged particles.

b) Identification and characterisation of natural products that may improve the learning and memory capability in Ts1Cje mouse model.

c) Exploration of the role of novel miRNAs in early embryo development.

d) Characterisation of candidate noncoding RNAs in human Down syndrome induced pluripotent stem cell (iPSC)-derived neurones.

e) Elucidation of long noncoding RNAs roles in gene expression regulation in the mammalian brain via bioinformatics and genomics approaches.

 

NeuroBiology and Genetics Group, UPM. (from left): Mr Mohd Hamdi Noor b. Abd Hamid, Mr. Wei-Hong Siew, Mr Alex Lee, Dr Michael KH Ling, Dr. Pike-See Cheah, Ms Angeline Leong, Ms Kai-Leng Tan and Ms Ameerah Jaafar.

Contact details:
Dr. Michael KH Ling: 
[email protected] +603-89472564
Dr Ling’s Scientific Malaysian profile: http://www.scientificmalaysian.com/scimy/members/michaelling/

Dr. Pike-See Cheah:
[email protected] +603-89472355
Dr Cheah’s Scientific Malaysian profile: http://www.scientificmalaysian.com/scimy/members/pikesee/

Links:
Research Cluster: http://medic1.upm.edu.my/jog/neuro/

Organics Electronics – Universiti Malaya

 

TOP LEFT: The world’s most energy efficient yellow organic light-emitting diode (OLED) made in our lab. BOTTOM LEFT: Blue OLED which is the foundation for white OLED, has efficiency almost as good as the best in the world. RIGHT: The OLED team members in the Low Dimensional Materials Research Center at University of Malaya.

TOP LEFT: The world’s most energy efficient yellow organic light-emitting diode (OLED) made in our lab. BOTTOM LEFT: Blue OLED which is the foundation for white OLED, has efficiency almost as good as the best in the world. RIGHT: The OLED team members in the Low Dimensional Materials Research Center at University of Malaya.

We are a group of chemists, engineers and physicists working in the field of organic electronics. Our mission is to bring organic electronics from the lab to pre- commercialisation stage. Our vision is to create a more environmentally friendly electronics at low cost and at the same time to encourage techno-entrepreneurship in Malaysia.

Our research focuses on efficiency and performance of the materials and devices. By studying the charge transport, interfacial barriers, energy levels of organic semiconducting materials, exciton confinement, exciton dissociation, synthesis of various novel organic semiconducting materials, molecular modelling and correlation between molecular structure and device properties, we hope to achieve devices with very high performance. We fabricate various devices such as organic field effect transistors, organic phototransistors, organic light emitting diode, organic photovoltaic and smart organic electronics. We also actively patent our new findings.

We are seeking collaborators with experience in organic synthesis, device modelling and integration of different devices into user-friendly products. We are also seeking industries which are interested to bring such technology into commercial world. 

Contact:
Dr. Woon Kai Lin:
E: [email protected] T: +603-79674287 A216, Department of Physics
Faculty of Science
50603, Kuala Lumpur
University Malaya
Malaysia

Links:
Dr. Woon’s Scientific Malaysian profile: http://www. scientificmalaysian.com/members/kailin/
Dr. Woon’s website: http://fizik.um.edu.my/wkl/