With data collected from about a dozen pilot-phase laboratories, leaders of the Human Brain Proteome Project expect to present preliminary results of their project, in the form of protein lists, at the 4th annual Human Proteome Organization congress scheduled to take place in Munich at the end of August
During the pilot phase of HBPP, samples from the brains of mice of three different developmental stages 16-day-old embryos, 7-days-old embryos, and 8-weeks-old embryos were sent out to about 20 laboratories worldwide. In addition, two human samples were sent out to the laboratories: a control, a post-mortem brain sample, and a sample from the brain of a living, epileptic patient.
About 12 laboratories returned data by the submission date in March.
Being presented at HUPO will be "some first protein lists in terms of overlapping proteins between the mice-developmental stages, and also large overall protein lists," said Christian Stephan, a group leader in the bioinformatics division at the Medical Proteom-Center at Ruhr University in Bochum, Germany, who is also in charge of the HBPP Data Collection Center in Bochum.
The HBPP differs from the Human Plasma Proteome Project, which presented preliminary results from its pilot phase during last year's HUPO congress in Beijing. In the HBPP, data was collected, stored, and submitted as mass spectrometry peak lists in Bruker Daltonics' ProteinScape data-collection software, instead of being submitted as protein lists after analysis by each individual laboratory.
The blinded HBPP data is currently being reprocessed and analyzed at the Data Collection Center in Bochum, in collaboration with researchers at the European Bioinformatics Institute in Hinxton, UK. The peak lists are run through four different search engines: Sequest, Mascot, GeneBio's Phenyx, and Protagen's PFF Solver.
"The reason we do it this way is that if some groups do their own analysis and we try to compare between different participating groups, it's very difficult to do that because the way they decide on the protein lists is totally different," said Stephan. "We take all the peak lists from all participating groups and do central reprocessing."
According to Stephan, researchers decided to use ProteinScape to store data for the HBPP because it is the only software that allows for the storage of 2-D gel, 1-D gel, 1-D liquid chromatography, and 2-D LC data. In addition, the software integrates well with each of the four different search engines, and it is easy to use and readily available as HTML/web-based software.
Aside from being different from HPPP in terms of data collection, the HBPP also differs from HPPP in terms of its protein-finding goals: while HPPP aimed to catalog all proteins in plasma, HBPP aims to find proteins that differ in abundance between different developmental stages of mice, and between diseased and healthy humans.
Two bioinformatics specialists Steve Stein of the National Institute for Standardization testing, and Eugene Kapp of the Ludwig Institute for Cancer Research in Parkville, Australia will be comparing data from the HBPP with data from the HPPP to see what comparisons can be made in terms of biomarkers and proteins expressed.
After analysis of the pilot phase is completed, HBPP members plan to move on to the main phase of the project, which will focus on comparing wild type mice with mouse models of neurodegenerative diseases to look for biomarkers of neurodegenerative diseases.
It is unclear when the main phase of HBPP will start, Stephan said.
In addition to analyzing brain tissue, the HBPP also includes a subproject called the Clinical Neuroprototeomics of Human Body Fluids. Following a kick-off meeting held last December at the Castle Rauischholzhausen in Germany, coordinators of the subproject plan to start sending out cortical spinal fluid samples to participating laboratories early next year.
"The subproject will focus directly on humans," said Stephan. "We want to analyze human bodily fluids to see if there are some special markers that can give hints for the onset of neurodegenerative diseases."
Following the HUPO congress in Munich, HBPP members plan to hold a data analysis jamboree in mid-October at the EBI in Hinxton.
"We're going to try to invite some specialists for analyzing this data," said Stephan.
In addition to analyzing differences in protein expression between developmental stages and disease versus normal states, researchers will also analyze where proteins were located within cells.
Another secondary focus of the analysis will be to see what differences there are that correlate to the use of different search engines and different technologies.
"It's not easy to say at this stage what differences there will be in terms of what is found by the different search engines, or between 2D-gels versus 2-D liquid chromatography. That's a secondary goal," said Stephan. "Our first goal is to look for differing protein amounts between the different time stages of the mice, and the different human samples."
Tien-Shun Lee ([email protected])