Yonsei Proteome Research Center in Seoul, South Korea
At A Glance
Name: Young-Ki Paik
Position: Director, Yonsei Proteome Research Center in Seoul, South Korea, since 2000; Professor and associate professor, department of biochemistry, Yonsei University, since 1993; President, K-HUPO, since 2001; Secretary General, HUPO, since 2001.
Background: Associate Professor, department of biochemistry, Hanyang University in Ansan, South Korea, 1989-1993.
Postdoc and staff research investigator, Gladstone Institute of Cardiovascular Diseases, University of California San Francisco, 1983-1989.
PhD in biochemistry, University of Missouri Columbia, 1983.
BS in biochemistry, Yonsei University, 1975.
Young-Ki Paik was elected to be one of the next vice presidents of HUPO, along with Catherine Fenselau of the University of Maryland, at the HUPO Fourth Annual World Congress held in Munich last week. ProteoMonitor caught up with Paik, who will begin his term as HUPO vice president in January 2007, to find out more about his accomplishments so far, and his views on HUPO's future.
You are one of the founding leaders of proteomics in South Korea. How did you get involved in the field and in establishing the Yonsei Proteome Research Center?
Way back in 1999, there was an opportunity for us to apply for a national grant, a big project under the name of functional genomics. So that gave us a moment to start thinking of proteomics, and because our department at Yonsei University has a long tradition of protein chemistry and protein biochemistry, from within functional genomics, we chose proteomics as a main workhorse for our research team.
At that time, I established the YPRC within Yonsei University, which consists of more than 20 faculty members, and about 50 graduate students. We didn't have that much experience at the time. However, we were able to learn more proteomics by collaborating with APAF the Australian Proteome Analysis Facility in Sydney and also by inviting world leaders to symposiums, and also hosting some proteomics workshops and by participating in many other proteomics meetings. Personally, I didn't learn proteomics before 1999, but after then I started learning and working on the proteins, and I've been working on many, many enzymes and especially membrane-bound proteins for about 20 years.
I was working on membrane-bound proteins way back in 1979, when I was purifiying cholesterol biosynthesis enzymes. It was very sticky. That gave us some feeling about proteomics. Proteomics requires a lot of global work more extensive speed and more high-throughput techniques.
So that's the story of how I started in proteomics. Since then, we were able to receive several grants which focus on standardization and bioinformatics, and also disease proteomics and the plasma proteome project in Korea. In 2003, we started a new project called the Korean Human Proteome Project. That money comes from the Minister of Health and Welfare which supports annually about $2.1million to $2.3 million in research, depending on the exchange rate.
When did you start working with mass spec and high performance liquid chromatography?
HPLC was quite a long time ago, since 1979, 1980. We used all different types of columns these were our basic tools for protein biochemistry. We added mass spec in 1999.
Now we have a very good level of techniques at YPRC which allows us to analyze all different types of peptides and samples. We are now in a very comfortable position to run any type of machine.
What were the first proteome projects that you worked on?
We started with liver cancer and stomach cancer, because those two cancers are major cancers in Korea. And one year later we added lung cancer in the line of our subject. We've done pretty well in terms of publication and constructing our infrastructure. We have some expertise in sample preparation, as well as profiling of all disease samples.
How many labs are involved in the projects?
Depends on the project. I run five different projects at the moment. So for the plasma proteome project, we call it K-HPPP Korean Human Plasma Proteome Project which is run in collaboration with HUPO's HPPP, there are a few faculty members actively involved in the project. One of our functions is to support 14 medical genomics centers in Korea, with respect to some educational training, data handling and technical services. Because they don't have any facilities to run their proteomic samples for a specific disease, they come to us, we analyze the samples and return the results to them for their publication or report.
We have two missions: Our core project is focused on the K-HPPP. Secondly, we aim to provide technical services and educational training for 14 medical genomics centers throughout the country.
We are also involved in all aspects of HPPP, which is run by Dr. Gilbert Omenn. For that project, we will have two more core leaders soon I'll be one of them, hopefully. The leaders will be finalized in the October meeting in Beijing, where we're going to have a roundtable discussion of all HUPO initiatives. Each director will identify their core leaders, and present some initiative proposal, and our HUPO executive council will review and finalize the proposal. So this is about two months, one and a half months away.
Our center has been specialized in liver cancer disease proteomics, so it is very obvious for us to also be involved in Human Liver Proteome Project. We are providing some data, especially some post-translational modification data from specific proteins.
Disease-wise, we also have done very good projects on lung cancer proteomics, which is focused on certain marker proteins in the context of their molecular and cellular function.
What's the next stage after finding biomarker proteins for the various cancers?
The next stage is validation works using different tools. We have a potential candidate for liver cancer and also lung cancer. We're going to use them for the extensive cellular and molecular biology studies.
Personally, I am also interested in some aging-related proteins from C. elegans, which is an excellent model for aging studies. We recently found a very interesting molecule called daumone, published in Nature in February. The molecule controls the extended life cycle of C. elegans. It induces a non-aging state so that the worms could lead a life 10 times or eight times longer than their original life.
What do you think is the advantage of having an international or national body of laboratories work on a single project?
There are obvious advantages for doing this kind of global project with an international entity like HUPO. First, the amount of data is huge. Compared to the genomics data, it's maybe 100 times that, because the proteins create a lot of spaces for data sets. Also, the data has many variations. Single genes usually make 10 to 20 protein products. Twenty proteins from a single gene can be present in a cell. To map all the proteins present in a disease state or healthy state, we need to identify what kinds of modifications are there and what kinds of structures. Some proteins are truncated. Some proteins are processed in different ways.
Also there is no standardized method of handling data and identification of proteins. There has to be a set of standard ways for validating the identified proteins. That hasn't been set yet. Throughout the Plasma Proteome Project, we realized there is an urgent need to standardize all types of data. Those standards could be accepted for publication, or for drug development. That's why we're working together.
There are also many areas of expertise needed to solve a single problem there are computer scientists, informaticians, biologists, clinicians, proteomics scientists and mass spec scientists all are needed to identify one single protein which has a certain specific function, structure and modification.
Also a protein has a lot of networks. Any single protein does not work by itself. It always interacts with some other proteins and networks to carry out a single function. Even if you look at one single enzyme catalyzing A to B, it doesn't work alone. It requires a lot of cofactors, certain other proteins, and then as a team it works as a cascade.
So to do high-throughput proteomics, we need to develop tools, like a microarray system and high-throughput screening system. We need to make a lot of antibodies that's why we recently announced Human Antibody Inititative led by Dr. Matthias Uhlen. We need core facilities and database systems, like the ones developed by the European Bioinformatics Institute.
So this requires a lot of effort from different teams and disciplines. That's why we need to coordinate in some organized fashion. That's why HUPO is directing this kind of project coordinating teams and focusing the project and making data accessible to the public. If one company does that on its own, they never would publish their results so that they are accessible to other groups. But HUPO is a non-profit organization, and it opens this kind of data to the public.
How does HUPO deal with all the different subgroups that want to become new initiatives of the organization?
Joining HUPO is on a totally voluntary basis. We identified good groups who have expertise in certain fields, and we let them propose their plans.
Samir Hanash is chair of the HUPO Initiatives Committee. He asks people that are interested in a type of global HUPO project to propose a draft plan. He lets them organize workshops internationally it can't just belong to a single country and lets them discuss, and choose their leadership. From HUPO's point of view, we look at whether or not they are mature enough to announce that they are going to be a HUPO project.
So there is a system. They will propose to the HUPO executive council, and at the HUPO council meeting, the members will decide if it's going to be a HUPO project or not.
If somebody wants to use the HUPO name for their own private project, it's not allowed. That wouldn't make sense. So we are setting up some ethical standards to prevent any possible conflict of interest.
What do you see changing within HUPO within the next few years?
From my personal point of view, HPPP certainly is the leading project, and I believe their pilot phase is finished. Gil Omenn will lead the HPPP to the execution phase, which requires more extensive work and highly defined teams. For any team that didn't have any product in the pilot phase, it's going to very difficult for them to joint the project in the execution phase. So there has to be some kind of credibility during the pilot phase.
During the pilot phase, we experienced a lot of variability in terms of HPPP teams. There were problems with data variability, data handling. In the execution phase, we will use those kinds of experiences to judge which teams could be good ones, which systems could be better ones, things like that.
What do you see coming up for K-HUPO in the next few years?
I'd like to mention that AO-HUPO Asia-Oceanic HUPO I'm secretary general of that organization has recently started thinking of our own project that we call Membrane Protein Initiative, or MPI. AO-HUPO will soon have a meeting to review a proposal drafted by Bill Jordan from New Zealand, and to go over key points of the project. It was a very nice draft plan to focus on the membrane-bound proteins, regardless of any animal models or cell types. So we are focusing on the technical know-how to profile all membrane bound proteins, which are very sticky proteins known to be very hard to analyze. About 13 countries have now joined AO-HUPO.
Korea certainly has moved forward in the advancement of proteomics, and also as a member of collaborative projects, such as HPPP, HLPP, HAI and the Human Brain Proteome Project. In terms of bioinformatics, I recently sent one postdoc fellow to European Bioinformatics Institute. He's going to learn and then manage some of the Korean data as well as HUPO data, and coordinate some big projects like PRIDE. He will make a good link between the Proteomics Standards Initiative and YPRC.
Our lab has been involved in many of the HUPO projects, and as a whole, the Korean team has a very good opportunity to be involved in the HUPO projects. I was told by the HBPP team that the Korean teams produced a lot of good, high-accuracy protein identification data. It was very impressive, and I'm very proud of it.
Are there a lot of Korean proteomics companies?
We have only two or three venture companies. They provide certain types of services, or they are specialized in certain types of protein arrays or antibodies. There is an antibody company called LabFrontier that is working together with Matthias Uhlen.
I'd like to mention finally that K-HUPO will sponsor the 2007 HUPO Congress in Seoul. I'll be the organizing chair, and HUPO officers will be the colleagues from AO-HUPO. I'm looking forward to it. We're working very hard to receive more than 3,000 attendees.