Keynote Speakers

Plants produce a wide variety of metabolites through chemical reactions catalyzed by a number of enzymes. Plant metabolism, which is comprised of primary and specialized (secondary) metabolisms, is finely regulated in response to internal and environmental stimuli and forms a complex network. As plant metabolites are important for not only plants but also human as nutrients, flavors, pigments, medicinal and health-promoting compounds, understanding plant metabolic reaction networks comprising of enzymatic reactions and their regulatory mechanisms is receiving attention for metabolic engineering and synthetic biology.

Over the past decade, it has become possible to acquire a large metabolome dataset from high-throughput technologies based on mass spectrometry and nuclear magnetic resonance. We consider time-series metabolome data especially important to understand metabolic reaction networks because they include in vivo kinetic information of networks. We are developing a new approach to integrate the time-series metabolome data with metabolic reaction network for building a large-scale kinetic model. In this approach, a mathematical model is constructed using power-law representation within the framework of Biochemical Systems Theory.The model parameters are simultaneously estimated using only actual time-series data of metabolite concentrations. The obtained model enables us to simulate dynamic behaviors of metabolite concentrations and perform system analysis of the network. In this presentation the concept and the results of our on-going study will be introduced.

References:

1. Hirai, M.Y., Yano, M., Goodenowe, D.B., Kanaya, S., Kimura, T., Awazuhara, M., Arita, M., Fujiwara, T., and Saito, K. (2004) Integration of transcriptomics and metabolomics for understanding of global responses to nutritional stresses in Arabidopsis thaliana.  Proc. Natl. Acad. Sci. U. S. A. 101, 10205-10210.
2. Sawada, Y., Akiyama, K., Sakata, A., Kuwahara, A., Otsuki, H., Sakurai, T., Saito, K., and Hirai, M.Y. (2009) Widely targeted metabolomics based on large-scale MS/MS data for elucidating metabolite accumulation patterns in plants. Plant Cell Physiol. 50, 37-47.
3. Sriyudthsak, K., Shiraishi, F., and Hirai, M.Y. (2013) Identification of a metabolic reaction network from time-series data of metabolite concentrations. PLoS One 8, e51212.
4. Sriyudthsak, K., Iwata, M., Hirai, M.Y., and Shiraishi, F. (2014) PENDISC: a simple method for constructing a mathematical model from time-series data of metabolite concentrations. Bull Math Biol. 76, 1333-1351.
5. Sriyudthsak, K., Sawada, Y., Chiba, Y., Yamashita, Y., Kanaya, S., Onouchi, H., Fujiwara, T., Naito, S., Voit, E.O., Shiraishi, F., and Hirai, M.Y. A U–system approach for predicting metabolic behaviors and responses based on an alleged metabolic reaction network. Accepted in GIW2014.

At the center of several wide-spread diseases is an evolutionary process. In infectious diseases, a foreign pathogen invades the human host and exploits it to produce progeny. The infected host stages an immune response targeted at exterminating the pathogen. In medical care, this process is supported with drug therapy (antibiotics or antiviral drugs). In response the pathogen evolves to forms that evade the immune system and are resistant to the applied drugs. Once resistance arises the drug therapy has to be changed appropriately to be effective against the newly evolved pathogenic strain. For some viruses, antiviral drugs are routinely given in combination in order to most effectively curb replication of the pathogen. The selection of a suitable drug combination rests on an analysis of the resistance profile of the current viral strain, is patient specific and is so complex that it requires computer support.

Over the past decade bioinformatics technology has advanced to effectively aid the selection of drug therapy in this setting. Forerunner is the HIV infection leading to AIDS, where bioinformatics-assisted therapy selection has entered clinical routine. Hepatitis C and Hepatitis B are following. Cancer has a similar evolutionary characteristic. Here, the role of the pathogen is taken by the aberrant tumor genome. Research in the cancer field is expected to follow in the tracks of the developments for viral diseases.

We report on the state of the art in the field of bioinformatics-supported resistance analysis and give perspectives of further developments.

Selected publications:

• Lengauer, T., Pfeifer, N. and Kaiser, R. (2014). “Personalized HIV therapy to control drug resistance.” Drug Discovery Today: Technologies 11:57-64.
• Bozek, K., et al. (2013). “Analysis of Physicochemical and Structural Properties Determining HIV-1 Coreceptor Usage.” PLoS Comput Biol 9(3): e1002977.
• Bock, C. and T. Lengauer (2012). “Managing drug resistance in cancer: lessons from HIV therapy.” Nat Rev Cancer 12(7): 494-501.
• Bogojeska, J. and T. Lengauer (2012). “Hierarchical Bayes model for predicting effectiveness of HIV combination therapies.” Stat Appl Genet Mol Biol 11(3): Article 11.
• Pfeifer, N. and T. Lengauer (2012). “Improving HIV coreceptor usage prediction in the clinic using hints from next-generation sequencing data.” Bioinformatics 28(18): i589-i595.

Over the past decade, many methods have been proposed to find relevant disease-causing mechanisms from gene-expression data. To date, no method is able to reliably identify disease mechanisms in extremely-small-sample-size situations. Even in a moderately-large-sample-size situation, gene-expression analysis shows low consistency when applied to independent datasets of the same disease phenotypes. In this talk, we first dissect the reasons for the failure of some well-known pathway-based gene-expression analysis methods. Surprisingly, these methods fail for various rather fundamental reasons, including inappropriate null hypothesis, instability when sample size is small, and signal dilution from normal-behaving branches of large relevant pathways. Then, by logically fixing each issue directly and rather simply, we show a quantum leap in the reproducibility, precision, and sensitivity of gene selection from gene-expression profiles, even when sample size is extremely small and even when there are significant batch effects in the sample.

The genomes of extinct hominins closely related to present-day humans offer a unique opportunity to identify those genetic changes specific to anatomically fully modern humans as well as providing insights into human population history. Using next-generation sequencing we have generated high quality genome sequences for two archaic hominin groups: Neandertals (1) who lived in Europe and Western Asia until approximately 30 000 years ago, and Denisovans (2), a group that was recently described based on the genome sequence generated from a bone found in Southern Siberia. We have also sequenced to high coverage the genome of a 45,000 year-old modern human from Siberia. The availability of these high coverage genomes provides multiple insights into the population history of both archaic and modern humans. Among other insights we have found that about 2.0% of the genomes of present-day non-Africans derive from Neandertals while about 4.8% of the genomes of present-day Oceanians derive from Denisovans. We have identified the regions of Neandertal introgression in present-day people and characterized some of the functional impacts of Neandertal ancestry on modern humans (3). Further we have identified sequence differences that have come to fixation or reached high frequency in modern humans since their divergence from Neandertals and Denisovans, some of which may have important functional effects in modern humans.
I will outline some of the challenges in the generation and analysis of ancient genome sequence data, and discuss the evolutionary insights that have resulted from the sequencing of these genomes.

References

1. K. Prufer, F. Racimo, N. Patterson et al. The complete genome sequence of a Neanderthal from the Altai Mountains. Nature 505, 43-49 (2014).
2. M. Meyer, M. Kircher, M. T. Gansauge et al. A high-coverage genome sequence from an archaic Denisovan individual. Science 338, 222-226 (2012).
3. S. Sankararaman, S. Mallick, M. Dannemann et al. Nature 507, 354-357 (2014).

Insulin regulates many metabolic functions, such as glycogenesis, gluconeogenesis and protein synthesis, through the AKT pathway. Blood insulin reportedly exhibits several temporal patterns, such as additional secretion, which is a pulse-like secretion in response to meals, and basal secretion, which is the low and constant secretion during fasting, suggesting that insulin selectively regulate downstream functions depending on its temporal patterns. To explore temporal coding of insulin action, we developed a simple computational model of the insulin-dependent AKT pathway, and found that the pathway uses “temporal patterns” for selective downstream regulation via differences in their network structures and kinetics (Kubota et al., Mol. Cell, 2012). Pulse and sustained insulin stimulations were simultaneously encoded into transient and sustained AKT phosphorylation, respectively. And downstream molecules, including S6K, G6Pase and GSK3β, selectively decoded transient, sustained, and both transient and sustained AKT phosphorylation, respectively. We also elucidated that the glucose metabolisms including glycolysis, gluconeogenesis and glycogenesis are selectively regulated by temporal change and absolute concentration of insulin (Noguchi et al., Mol. Sys. Biol., 2013). Our results demonstrate that the AKT pathway can multiplex distinct patterns of insulin and the downstream molecules selectively decode the temporal patterns of insulin.

The fast progression of genomics is making of the use of personal genomic information a pressing daily reality. In this scenario Bioinformatics and Computational Biology play a central rôle. The organization and analysis of individual genomes is a complex task involving data organization, integration and interpretation challenges, and requires a blend of engineering and scientific developments at each step of the analysis, since it touches many areas in which the development of computational methods is far from complete.In the context of various (epi)genome projects, my group is developing both the technical framework for handling the data and the methods required for the interpretation of the information. Based on our experience in these projects, I will review some of the key problems in the analysis of high-throughput genomic information, including the prediction of the incidence of mutations with special emphasis in the application of co-evolution related methods, the implications of the alterations of the splicing machinery, and the comparative analysis of disease affected pathways.

References

1. Valencia A & Hidalgo M (2012) Getting personalized cancer genome analysis into the clinic: the challenges in Bioinformatics. Genome Medicine, 461
2. Vazquez M, de la Torre V , Valencia A (2012) Chapter 14: Cancer Genome Analysis, in Translational Bioinformatics PLOS Computational Biology open access book.
3. de Juan D, Pazos F, Valencia A. (2013) Emerging methods in protein co-evolution. Nat Rev Genet. 14:249-61.
4. Rodriguez JM, Maietta P, Ezkurdia I, Pietrelli A, Wesselink JJ, Lopez G, Valencia A, Tress ML. (2013) APPRIS: annotation of principal and alternative splice isoforms. Nucleic Acids Res.  41:D110-7.
5. Ibañez C, Boullosa C, Tabarés-Seisdedos R, Baudot A and Valencia A (2014) Molecular Evidence for the Inverse Comorbidity between Central Nervous System Disorders and Cancers detected by Transcriptomic Meta-analyses. Plos Genet, in press.