A probabilistic view of gene function

Abstract

Cells are controlled by the complex and dynamic actions of thousands of genes. With the sequencing of many genomes, the key problem has shifted from identifying genes to knowing what the genes do; we need a framework for expressing that knowledge. Even the most rigorous attempts to construct ontological frameworks describing gene function (e.g., the Gene Ontology project) ultimately rely on manual curation and are thus labor-intensive and subjective. But an alternative exists: the field of functional genomics is piecing together networks of gene interactions, and although these data are currently incomplete and error-prone, they provide a glimpse of a new, probabilistic view of gene function. We outline such a framework, which revolves around a statistical description of gene interactions derived from large, systematically compiled data sets. In this probabilistic view, pleiotropy is implicit, all data have errors and the definition of gene function is an iterative process that ultimately converges on the correct functions. The relationships between the genes are defined by the data, not by hand. Even this comprehensive view fails to capture key aspects of gene function, not least their dynamics in time and space, showing that there are limitations to the model that must ultimately be addressed.The necessity for a logical framework A primary issue in biology is describing the global organization of genes into systems and understanding the coordination of these systems in the cell. The sequencing of complete genomes has yielded the full parts lists of several organisms, and we now have powerful techniques to sample different aspects of gene function at a genome-wide level. Together, these advances hold great promise in leading us closer to a complete description of the molecular biology of the cell. For us to reach this goal, however, we need some coherent framework in which to express what we learn about gene function through the accumulation of data. What do we mean by 'gene function'? More importantly, what should we mean by this? What are the key properties of genes that we should measure to adequately describe their function in the cell? If we have all the measurements, how should we integrate them into a complete description of the molecular machineries that are the basis of a cell? Our framework for thinking about gene function inevitably colors the questions we ask and the conclusions we draw; therefore, defining a rational framework is not merely an exercise in abstraction but a key tool in understanding how an organism works.Great effort (e.g., refs. 1-8) has already gone into defining precisely what we mean by the functions of genes, as well as how we should record this information. These summaries help us by identifying common features of genes with similar functions and giving clues to the organization and control of processes in the cell. They may also help us to see whether this organization can explain observed biological properties and guide us toward new processes. The functional framework itsel...

Widespread reorganization of metabolic enzymes into reversible assemblies upon nutrient starvation

“…For example, both the ribosome and tRNA multisynthetase complex (MSC) have been found to harbor components that can migrate away from the complex and in so doing acquire or regulate new functions (1). In addition, the extraordinary interconnectivity of protein interactome networks strongly suggests that many genes have multiple functions, and indeed that gene function in general can be considered to be ''probabilistic,'' with each protein dedicated partially to a number of distinct tasks (2). Between these findings, it seems likely that many proteins will be both spatially and functionally organized into dynamic complexes that could assemble and disassemble dependent upon the needs of the cell (3).…”

Predicting genetic modifier loci using functional gene networks

“…In biology, however, functional relationships among proteins often transcend direct physical interactions (Fraser and Marcotte 2004). Many proteins can be important for common biological processes without physically interacting or associating.…”

Detecting remotely related proteins by their interactions and sequence similarity

“…The first group of methods includes sequence matching and threading (9,11). The second group includes both experimental and computational methods, such as clustering by physical interactions (26), mRNA array expression profiles (38), analysis of gene fusion (30), phylogenetic profiles (39), and genomic association of genes (40). Our approach is unique in that it discovers homology by explicitly combining both sequence similarity and experimentally determined protein interactions.…”

“…One such special case are the ''interlogs'' (i.e., pairs of interacting proteins that interact identically in two species) (19). It has already been demonstrated that the information about the interacting partners can be used to predict the fold (7,(20)(21)(22) or function (14,(23)(24)(25)(26)(27)) of a protein without considering its sequence. The usefulness of these approaches should grow with time, given the increasing amount of data about protein-protein interactions (20,24,28) (4).…”