Talk title: Knowledge Mining from -OMICS Datasets
Fuchu He, Member of Chinese Academy of Sciences and Academy of Sciences for the Developing World, is the leading scientist studying proteomics in China. He was among the first group of people who founded HUPO in 2001 and the founder of Chinese arm of HUPO (CNHUPO). He was the first Chinese scientist who led an international consortium to undertake a large-scale project to decipher the human liver proteome as the inaugural chair of Human Liver Proteome Project (HLPP), the first proteome project for human organs, which remarkably applauded by Nature and Science. Now as the chief scientist, he is propelling HLPP to the second step, China Human Proteome Project(CNHPP), to create an encyclopedia of proteins in the human body under physiological and pathological conditions. He founded Beijing Proteome Research Center, State Key Laboratory of Proteomics and National Center for Protein Sciences∙Beijing-PHOENIX (Pilot Hub Of ENcyclopediac proteomIX), and is currently their president. He was the Inaugural President of Institutes of Biomedical Sciences, Fudan University.
His major fields of research are genomics, proteomics, bioinformatics and systems biology. To date, he has published more than 270 papers in international peer-reviewed journals, such as Nature and Science. Now he is taking a systems biology approach to uncover evolutionary rules governing proteomes and to understand their misregulation leading to disease.
1State Key Laboratory of Proteomics, Beijing Proteome Research Center, Beijing Institute of Radiation Medicine;
2National Center for Protein Sciences∙Beijing (Pheonix Center), Beijing 102206, China
Throughout the history of natural science, it is definite that the discovery of our knowledge and disciplines are triggered by the unprecedented scale and speed of big data and achieved by efficient mathematical strategies.
In the past 25 years, mathematical strategies have been used in my laboratory to generate multiple biological findings based on large-scale datasets. The story began with simple statistical methods used to find four periodic phenomena of molecular evolution in the early years.
Next, machine learning strategies, such as naïve Bayesian network, have been used to find the instinct features of proteome organization, especially the protein interactions.
At the current stage, clustering strategies are playing important roles for the molecular characteristics of HCC and new personalized treatment strategies based on large scale human proteome datasets.
The era of big data will bring in new insights in life sciences and present new opportunities in research. Artificial intelligence strategy will play dominant roles in the coming knowledge discovery. My team is now engaged in developing an automatic knowledge discovery highway for grand knowledge.