The recent proliferation of protein interaction networks has motivated research into network alignment: the cross-species comparison of conserved functional modules. Previous studies have laid the foundations for such comparisons and demonstrated their power on a select set of sparse interaction networks. Recently, however, new computational techniques have produced hundreds of predicted interaction networks with interconnection densities that push existing alignment algorithms to their limits. To find conserved functional modules in these new networks, we have developed Graemlin, the first algorithm capable of scalable multiple network alignment. Graemlin's explicit model of functional evolution allows both the generalization of existing alignment scoring schemes and the location of conserved network topologies other than protein complexes and metabolic pathways. To assess Graemlin's performance, we have developed the first quantitative benchmarks for network alignment, which allow comparisons of algorithms in terms of their ability to recapitulate the KEGG database of conserved functional modules. We find that Graemlin achieves substantial scalability gains over previous methods while improving sensitivity.

Abstract. We developed Graemlin 2.0, a new multiple network aligner with (1) a novel scoring function that can use arbitrary features of a multiple network alignment, such as protein deletions, protein duplications, protein mutations, and interaction losses; (2) a parameter learning algorithm that uses a training set of known network alignments to learn parameters for our scoring function and thereby adapt it to any set of networks; and (3) an algorithm that uses our scoring function to find approximate multiple network alignments in linear time.We tested Graemlin 2.0's accuracy on protein interaction networks from IntAct, DIP, and the Stanford Network Database. We show that, on each of these datasets, Graemlin 2.0 has higher sensitivity and specificity than existing network aligners. Graemlin 2.0 is available under the GNU public license at http://graemlin.stanford.edu.

We developed Graemlin 2.0, a new multiple network aligner with (1) a new multi-stage approach to local network alignment; (2) a novel scoring function that can use arbitrary features of a multiple network alignment, such as protein deletions, protein duplications, protein mutations, and interaction losses; (3) a parameter learning algorithm that uses a training set of known network alignments to learn parameters for our scoring function and thereby adapt it to any set of networks; and (4) an algorithm that uses our scoring function to find approximate multiple network alignments in linear time. We tested Graemlin 2.0's accuracy on protein interaction networks from IntAct, DIP, and the Stanford Network Database. We show that, on each of these datasets, Graemlin 2.0 has higher sensitivity and specificity than existing network aligners. Graemlin 2.0 is available under the GNU public license at http://graemlin.stanford.edu.

We have experimentally and computationally defined a set of genes that form a conserved metabolic module in the α-proteobacterium Caulobacter crescentus and used this module to illustrate a schema for the propagation of pathway-level annotation across bacterial genera. Applying comprehensive forward and reverse genetic methods and genome-wide transcriptional analysis, we (1) confirmed the presence of genes involved in catabolism of the abundant environmental sugar myo-inositol, (2) defined an operon encoding an ABC-family myo-inositol transmembrane transporter, and (3) identified a novel myo-inositol regulator protein and cis-acting regulatory motif that control expression of genes in this metabolic module. Despite being encoded from non-contiguous loci on the C. crescentus chromosome, these myo-inositol catabolic enzymes and transporter proteins form a tightly linked functional group in a computationally inferred network of protein associations. Primary sequence comparison was not sufficient to confidently extend annotation of all components of this novel metabolic module to related bacterial genera. Consequently, we implemented the Graemlin multiple-network alignment algorithm to generate cross-species predictions of genes involved in myo-inositol transport and catabolism in other α-proteobacteria. Although the chromosomal organization of genes in this functional module varied between species, the upstream regions of genes in this aligned network were enriched for the same palindromic cis-regulatory motif identified experimentally in C. crescentus. Transposon disruption of the operon encoding the computationally predicted ABC myo-inositol transporter of Sinorhizobium meliloti abolished growth on myo-inositol as the sole carbon source, confirming our cross-genera functional prediction. Thus, we have defined regulatory, transport, and catabolic genes and a cis-acting regulatory sequence that form a conserved module required for myo-inositol metabolism in select α-proteobacteria. Moreover, this study describes a forward validation of gene-network alignment, and illustrates a strategy for reliably transferring pathway-level annotation across bacterial species.

The collection of multiple genome-scale datasets is now routine, and the frontier of research in systems biology has shifted accordingly. Rather than clustering a single dataset to produce a static map of functional modules, the focus today is on data integration, network alignment, interactive visualization and ontological markup. Because of the intrinsic noisiness of high-throughput measurements, statistical methods have been central to this effort. In this review, we briefly survey available datasets in functional genomics, review methods for data integration and network alignment, and describe recent work on using network models to guide experimental validation. We explain how the integration and validation steps spring from a Bayesian description of network uncertainty, and conclude by describing an important near-term milestone for systems biology: the construction of a set of rich reference networks for key model organisms.