Severe Acute Respiratory Syndrome

When an outbreak of SARS hit Hong Kong in March 2003 the Centre diverted relevant human resources to fight the disease and to promote exchange with the World Health Organisation. A research programme will be set up in the domain for the next three years.

For a regular update see the Journalist page.

Training and Research Programme (2003-2005)


Details of infrastructure

1. Introduction

Hong Kong is looking forward to become the third Genome Centre in China. At the present time this no longer requires huge facilities in terms of sequencing (in fact the world-wide equipment in the domain is so large that many Centres are now looking for sequencing projects to avoid remaining idle). In contrast genome annotation and management facilities are more than ever needed. It was therefore important to bring Hong Kong and the region to the state-of-the-art knowledge and services in this domain, in a short period of time. This was the more important today because the international situation makes the Internet connectivity less secure and less reliable (connections are likely to be severed for a variety of reasons, including software attacks by all kinds of viruses or worms). The programme BIOSUPPORT, financed by the Innovation and Technology Fund, seeks to provide elements for a stable basis for in silico (computer mediated) genome studies in Hong Kong, which will be used by the Hong Kong community at large for all types of genome studies. In addition, if the appropriate support can be granted, this service aims at including Hong Kong in a network of international interactions which will allow scientists of the HKSAR, both academics and in private companies, to be part of the leading teams in the domain.

2. Background

Two broad conclusions from analysis of international literature and of workshops dedicated to the future of genomics gave rise to the scientific programme BIOSUPPORT aiming at providing Hong Kong with facilities which may be used as a help to discovery. First, it is clear from the presentations and discussions that functional genomics encompasses an enormously wide area and that there would be many benefits from an integrated approach between the experimental technologies and informatics tools, which are highly complementary in nature. Second, the combination of genome sequencing projects and functional analysis is producing a flow of data which puts a particular focus on the biological relationships between sequences; hence, the need for models, methods and tools to manipulate sequence networks or clusters is becoming ever more important. In this programme, our aims are to extract knowledge from the individual experimental "wet" technologies with the in silico methods which can be applied to them and to establish an appropriate infrastructure through which to optimise the exploration and exploitation of the enormous amount of information available, aiming in particular at the identification of idiosyncrasies of cells, organs or organisms, which make them differ from each other.

Many deficiencies and bottlenecks affect the high throughput functional genomics technologies in general. To begin with, there is a lack of software tools for data handling and of a widely accepted suite of analysis methods integrated into a coherent system for prediction of gene and protein stucture and function. In fact much software exist, but dispered at many sites, and, when they exist, packages are very expensive: it was therefore important to provide the region of Hong Kong and Southern China with packages which were both integrated and free to use. Secondly, there is an increasing problem of insufficient or erroneous functional assignment in sequence databases; part of the error propagation is caused by the lack of traceability of results, but most of the difficulties come from the lack of understanding appropriate model organisms: it is impossible to use a sequence such as the human genome sequence without having first explored experimentally appropriate models. Bacteria and unicellular eucaryotes remain the best tools in this respect, remembering that the most efficient drug targets aim at interfering with metabolic processes, rather than with the elaborate regulation cascade which are usually fashionable and studied throughout the world. Finally, naturally, large scale analyses cannot be performed on the World Wide Web, not only because reference services would immediately be clogged by large batches coming from all over the world, but above all because this makes protection of intellectual property impossible. Indeed connection to a remote service makes the service provider able to explore any questions submitted to it.

BIOSUPPORT seeks to put into place the necessary organisation to enable these problems to be tackled. The combination and integration of information which is currently dispersed and heterogeneous will create a general benefit through greater access to, and manipulation of, biological knowledge, including that in the literature as well as in databases. We believe that this programme will thereby facilitate the ultimate goal of functional genomics, which is the improved understanding of cell organisation at different levels, from individual genes to groups of biomolecules and complete genomes. Finally, as an example of an immediate consequence of this knowledge, identification and selection of appropriate drug targets will become possible, under conditions which are not usually explored, a necessary preliminary to the inventive activity needed for intellectual property protection.

The preferred organisation of BIOSUPPORT is simple. On the one hand, the HKU-Pasteur Research Centre is setting up, through its own activity and its collaborations in Hong Kong, China and Europe, the basic conditions needed to extract as much information as possible from genome studies. This work is validated by predictions which are tested locally as well as by a core of users which ensure by "alpha-testing" that what is proposed to the community is user-friendly. The second step of this programme (which has already developed with the creation of general core facilities), is providing the community with updated software packages and databases, through the Hong Kong University Computer Centre. This will eventually permit to set up a specialized node of the EMBNet structure, under the helm of the Bioinformatics Centre (the China EMBNet node) in Beijing. Naturally it is not possible to offer, at this stage, all the facilities present in the world. The program aims at providing the basic core on which everything can be further built up.

3. Services provided

Bioinformatics Tools:   The sequence analysis packages installed include EMBOSS, NCBI tools, FASTA, STADEN, PHYLIP, READSEQ, ClustalW/ClustalX, DIALIGN2 and the PHRAP/PHRED/CONSED package. Some tools installed also have on-line web interface, such as JEMBOSS, EMBOSS-GUI, NCBI-BLAST, FASTA and GenoList (Bacillus subtilis: SubtiList).

4. Working seminar in conceptual biology

In parallel with this development of a service for the Hong Kong region, a working seminar in conceptual biology has been ongoing on a weekly basis, in collaboration with the Department of Mathematics. This seminar, which has is totally open in its orientations, belongs to the "causeries du jeudi" series created by the Stanislas Noria group. It associates developments in mathematics, computer sciences, epistemology and anthropology, in collaboration with a network of universities in China and in Europe.

To be further documented

Understanding some of the Penicillium marneffei genome

1. Introduction

Dimorphic fungi, such as Histoplasma capsulatum, the most prevalent cause of fungus related diseases in man, develop as molds at relatively low temperatures, while they have a yeast form at the body temperature. In Asia another dimorphic fungus, Penicillium marneffei causes respiratory, skin and systemic mycosis. It is the most prevalent fungus causing such diseases in South East Asia. Discovered by a Pastorian scientist in 1956 in an hepatic abscess of the Chinese bamboo rat (Rhizomus sinensis), only 18 cases of human diseases were reported (in HIV negative patients) until 1985. Subsequently the disease exploded in HIV positive patients who resided in or had travelled to South East Asia. Today, 10% of AIDS patients in Hong Kong are infected with this fungus. Penicilliosis marneffei is an intracellular infection of the reticulo-endothelial cells by the yeast phase of the fungus. Despite the medical importance and its unusual dimorphic capability, studies of this fungus at the molecular level have been scarce. Only one cell wall mannoprotein gene has been characterised and sucessfully used in serodiagnosis of this infection. Based on the mitochondrial and spacer rRNA which allowed investigators to suggest a strong phylogenetic connection with Talaromyces sp., a PCR / hybridisation assay was designed for molecular identification of this fungus in positive cultures. Penicillium marneffei lacks a known sexual stage. A comprehensive knowledge of the genome would enable investigators to understand the basic mechanisms of dimorphism, disease pathogenesis, virulence, and immune defence.

2. Genome sequence tags

Several thousand sequence tags have been sequenced at the Institut Pasteur, from a strain isolated in a patient at the University of Hong Kong Faculty of Medicine. A preliminary work, developed in 2001, showed that one could probably distinguish between the genes specific of the mold stage, as compared to the yeast stage, according to their biases in using the genetic code. Further work showed that, in contrast to the case of many Candida species, it was likely that Penicillium marneffei uses the universal genetic code.

To be further documented

Research Programme (2000-2002)