The HUPO World Congress at Long Beach, 2006

This year’s HUPO World Congress, Bench to Bedside, captured the excitement within the proteomics community over current developments and the anticipation of applying these developments to fundamental and clinical questions. This was best reflected in the organizers’ intention that the Long Beach meeting was not going to be “just another mass spectrometry meeting.” Vertical integration of the program captured the breadth of developments in the fast moving proteomics field and illustrated the power of current technologies to better understand both basic biological functions and the pathogenesis to disease phenotypes. The opening keynote address by Dr. Anna Barker highlighted the congress theme on translating proteomics technologies to advance medicine and to benefit society. The breadth of research featured in this congress is best illustrated by two lectures presented at the meeting. Mike Tyers encapsulated the breadth of modern technologies applied to the protein interactome, and Donald Hunt described new developments in mass spectrometry technologies that can be used to better understand regulation of transcription. The network of 12,000 protein interactions in yeast curated from the primary literature (see BioGRID (General Repository for Interaction Datasets), www.thebiogrid.org) forms a dense, uniform network that contains more hub-hub interactions than the sparsely connected networks derived from initial high throughput studies. Recent high throughput (HTP) 1 proteome-wide surveys by the Greenblatt and Superti-Furga groups support the notion that biological networks are highly interconnected. Similarly the exhaustive analysis of the phagosome presented by Michel Desjardins and of the nuclear pore presented by Brian Chait supported the notion that biological processes are controlled by a dense and dynamic web of protein interactions. An important discussion point raised by John Yates, namely the reliability of individual interactions in an HTP dataset versus the same interaction studied in depth “by a postdoc at the bench” elicited divergent views from the HTP and small scale camps. Regardless it is clear that the yeast interactome is far from complete as, for example, less than 30% of literature protein interactions are found in all HTP data combined. Tyers also cited examples of well studied signaling pathways for which less than 5% of known interactions are found in an HTP dataset. Reinterrogation of the protein interactome using rapid magnetic bead-based isolation methods developed by Brian Chait holds promise of detecting the transient signaling interactions that underlie cellular dynamics. Given the evident density and redundancy of biological networks, Tyers argued that systematic collections of small molecules will be needed to precisely modulate the proteome. An initial matrix of 4,000 chemical-genetic interactions has enabled successful prediction of small moleculesmall molecule interactions. Such interactions may enable highly specific antifungal and anticancer therapeutics.