Motivation: Protein-protein interactions (PPIs) are usually modelled as networks. These networks have extensively been studied using graphlets, small induced subgraphs capturing the local wiring patterns around nodes in networks. They revealed that proteins involved in similar functions tend to be similarly wired. However, such simple models can only represent pairwise relationships and cannot fully capture the higher-order organization of protein interactions, including protein complexes. Results: To model the multi-sale organization of these complex biological systems, we utilize simplicial complexes from computational geometry. The question is how to mine these new representations of PPI networks to reveal additional biological information. To address this, we define simplets, a generalization of graphlets to simplicial complexes. By using simplets, we define a sensitive measure of similarity between simplicial complex network representations that allows for clustering them according to their data types better than clustering them by using other state-of-the-art measures, e.g., spectral distance, or facet distribution distance. We model human and baker's yeast PPI networks as simplicial complexes that capture PPIs and protein complexes as simplices. On these models, we show that our newly introduced simplet-based methods cluster proteins by function better than the clustering methods that use the standard PPI networks, uncovering the new underlying functional organization of the cell. We demonstrate the existence of the functional geometry in the PPI data and the superiority of our simplet-based methods to effectively mine for new biological information hidden in the complexity of the higher order organization of PPI networks.
Many proteins remain functionally unannotated. Sequence alignment (SA) uncovers missing annotations by transferring functional knowledge between species' sequence-conserved regions. Because SA is imperfect, network alignment (NA) complements SA by transferring functional knowledge between conserved biological network, rather than just sequence, regions of different species. Existing NA assumes that it is topological similarity (isomorphic-like matching) between network regions that corresponds to the regions' functional relatedness. However, we recently found that functionally unrelated protei…
Motivation: Protein interactions are fundamental building blocks of biochemical reaction systems underlying cellular functions. The complexity and functionality of such systems emerge not from the protein interactions themselves but from the dependencies between these interactions. Therefore, a comprehensive approach for integrating and using information about such dependencies is required. Results: We present an approach for endowing protein networks with interaction dependencies using propositional logic, thereby obtaining protein hypernetworks. First we demonstrate how this framework straig…
Directed protein networks with only a few thousand of nodes are rather complex and do not allow to extract easily the effective influence of one protein to another taking into account all indirect pathways via the global network. Furthermore, the different types of activation and inhibition actions between proteins provide a considerable challenge in the frame work of network analysis. At the same time these protein interactions are of crucial importance and at the heart of cellular functioning. We develop the Google matrix analysis of the protein-protein network from the open public database …
Are "turn-on" and "turn-off" functions in protein-protein interaction networks exact opposites of each other? To answer this question, we implement a minimal model for the evolution of functional protein-interaction networks using a sequence-based mutational algorithm, and apply the model to study neutral drift in networks that yield oscillatory dynamics. We study the roles of activators and deactivators, two core components of oscillatory protein interaction networks, and find a striking asymmetry in the roles of activating and deactivating proteins, where activating proteins tend to be syner…
Discovering patterns in networks of protein-protein interactions (PPIs) is a central problem in systems biology. Alignments between these networks aid functional understanding as they uncover important information, such as evolutionary conserved pathways, protein complexes and functional orthologs. The objective of a multiple network alignment is to create clusters of nodes that are evolutionarily conserved and functionally consistent across all networks. Unfortunately, the alignment methods proposed thus far do not fully meet this objective, as they are guided by pairwise scores that do not u…
One of the most difficult problems difficult problem in systems biology is to discover protein-protein interactions as well as their associated functions. The analysis and alignment of protein-protein interaction networks (PPIN), which are the standard model to describe protein-protein interactions, has become a key ingredient to obtain functional orthologs as well as evolutionary conserved pathways and protein complexes. Several methods have been proposed to solve the PPIN alignment problem, aimed to match conserved subnetworks or functionally related proteins. However, the right balance betw…
Biological networks provide insight into the complex organization of biological processes in a cell at the system level. They are an effective tool for understanding the comprehensive map of functional interactions, finding the functional modules and pathways. Reconstruction and comparative analysis of these networks provide useful information to identify functional modules, prioritization of disease causing genes and also identification of drug targets. The talk will consist of two parts. I will discuss several methods for protein-protein interaction network alignment and investigate their pr…
PPI network alignment aims to find topological and functional similarities between networks of different species. Several alignment approaches have been proposed. Each of these approaches relies on a different alignment method and uses different biological information during the alignment process such as the topological structure of the networks and the sequence similarities between the proteins, but less of them integrate the functional similarities between proteins. In this context, we present our algorithm PPINA (Protein-Protein Interaction Network Aligner), which is an extension of the NET…
How do living cells achieve sufficient abundances of functional protein complexes while minimizing promiscuous non-functional interactions? Here we study this problem using a first-principle model of the cell whose phenotypic traits are directly determined from its genome through biophysical properties of protein structures and binding interactions in crowded cellular environment. The model cell includes three independent prototypical pathways, whose topologies of Protein-Protein Interaction (PPI) sub-networks are different, but whose contributions to the cell fitness are equal. Model cells ev…
Genetic interactions have been widely used to define functional relationships between proteins and pathways. In this study, we demonstrated that yeast synthetic lethal genetic interactions can be explained by the genetic interactions between domains of those proteins. The domain genetic interactions rarely overlap with the domain physical interactions from iPfam database and provide a complementary view about domain relationships. Moreover, we found that domains in multidomain yeast proteins contribute to their genetic interactions differently. The domain genetic interactions help more precise…
We analyze large systems of interacting proteins, using techniques from the non-equilibrium statistical mechanics of disordered many-particle systems. Apart from protein production and removal, the most relevant microscopic processes in the proteome are complex formation and dissociation, and the microscopic degrees of freedom are the evolving concentrations of unbound proteins (in multiple post-translational states) and of protein complexes. Here we only include dimer-complexes, for mathematical simplicity, and we draw the network that describes which proteins are reaction partners from an en…
Sequence alignment forms the basis of many methods for functional annotation by phylogenetic comparison, but becomes unreliable in the `twilight' regions of high sequence divergence and short gene length. Here we perform a cross-species comparison of two herpesviruses, VZV and KSHV, with a hybrid method called graph alignment. The method is based jointly on the similarity of protein interaction networks and on sequence similarity. In our alignment, we find open reading frames for which interaction similarity concurs with a low level of sequence similarity, thus confirming the evolutionary rela…
Background: If biology is modular then clusters, or communities, of proteins derived using only protein interaction network structure should define protein modules with similar biological roles. We investigate the link between biological modules and network communities in yeast and its relationship to the scale at which we probe the network. Results: Our results demonstrate that the functional homogeneity of communities depends on the scale selected, and that almost all proteins lie in a functionally homogeneous community at some scale. We judge functional homogeneity using a novel test and th…
Cellular function is widely believed to be organized in a modular fashion. On all scales and at all levels of complexity, relatively independent sub-units perform relatively independent sub-tasks of biological function. This functional modularity must be reflected in the topology of molecular networks. But how a functional module should be represented in an interaction network is an open question. In protein-interaction networks (PIN), one can identify a protein-complex as a module on a small scale, i.e. modules are understood as densely linked, resp. interacting, groups of proteins, that are …
Vector Institute for AI Research（向量人工智能研究所）；University of Toronto（多伦多大学）
Protein function is inherently linked to its localization within the cell, and fluorescent microscopy data is an indispensable resource for learning representations of proteins. Despite major developments in molecular representation learning, extracting functional information from biological images remains a non-trivial computational task. Current state-of-the-art approaches use autoencoder models to learn high-quality features by reconstructing images. However, such methods are prone to capturing noise and imaging artifacts. In this work, we revisit deep learning models used for classifying m…
The architecture of the network of protein-protein physical interactions in Saccharomyces cerevisiae is exposed through the combination of two complementary theoretical network measures, betweenness centrality and `Q-modularity'. The yeast interactome is characterized by well-defined topological modules connected via a small number of inter-module protein interactions. Should such topological inter-module connections turn out to constitute a form of functional coordination between the modules, we speculate that this coordination is occurring typically in a pair-wise fashion, rather than by way…
Understanding functional associations among genes discovered in sequencing projects is a key issue in post-genomic biology. However, reliable interpretation of the protein interaction data has been difficult. In this work, we show that if two proteins share significantly larger number of common interaction partners than random, they have close functional associations. Analysis of publicly available data from Saccharomyces cerevisiae reveals more than 2800 reliable functional associations, 29% of which involve at least one unannotated protein. By further analyzing these associations, we derive …
Many important cellular functions are implemented by protein complexes that act as sophisticated molecular machines of varying size and temporal stability. Here we demonstrate quantitatively that protein complexes in the yeast, Saccharomyces cerevisiae, are comprised of a core in which subunits are highly coexpressed, display the same deletion phenotype (essential or non-essential) and share identical functional classification and cellular localization. This core is surrounded by a functionally mixed group of proteins, which likely represent short-lived- or spurious attachments. The results al…
A fundamental question for evolutionary biology is why rates of evolution vary dramatically between proteins. Perhaps surprisingly, it is controversial how much a protein's functional importance affects its rate of evolution. In most studies, functional importance has been measured on the coarse scale of protein knock-outs, while evolutionary rate has been measured on the fine scale of amino acid substitutions. Here we introduce dynamical influence, a finer measure of protein functional importance. To measure dynamical influence, we first use detailed biochemical models of particular reaction …
In this paper, we consider the statistical analysis of a protein interaction network. We propose a Bayesian model that uses a hierarchy of probabilistic assumptions about the way proteins interact with one another in order to: (i) identify the number of non-observable functional modules; (ii) estimate the degree of membership of proteins to modules; and (iii) estimate typical interaction patterns among the functional modules themselves. Our model describes large amount of (relational) data using a relatively small set of parameters that we can reliably estimate with an efficient inference algo…
Protein-protein interaction (PPI) networks are the backbone of all processes in living cells. In this work we relate conservation, essentiality and functional repertoire of a gene to the connectivity $k$ (i.e., the number of interaction links) of the corresponding protein in the PPI network. On a set of 42 bacterial genomes of different sizes, and with reasonably separated evolutionary trajectories, we investigate three issues: i) whether the distribution of connectivities changes between PPI subnetworks of essential and nonessential genes; ii) how gene conservation, measured both by the evolu…
Popular online enrichment analysis tools from the field of molecular systems biology provide users with the ability to submit their experimental results as gene sets for individual analysis. Such queries are kept private, and have never before been considered as a resource for integrative analysis. By harnessing gene set query submissions from thousands of users, we aim to discover biological knowledge beyond the scope of an individual study. In this work, we investigated a large collection of gene sets submitted to the tool Enrichr by thousands of users. Based on co-occurrence, we constructed…
The promise of discovering a functional blueprint of a cellular system from large-scale and high-throughput sequence and experimental data is predicated on the belief that the same top-down investigative approach that proved successful in other biological problems (e.g. DNA sequencing) will be as effective when it comes to inferring more complex intracellular processes. The results in this paper address this fundamental issue in the specific context of transcription regulatory networks. Although simple recurring regulatory motifs have been identified in the past, due to the size and complexity…
Mitochondrial diseases are largely caused by dysfunction in mitochondrial proteins. However, annotations of human mitochondrial proteins are scattered across various public databases and individual studies. To facilitate research aimed at elucidating mitochondrial functions, we constructed the MEGADOCK-Web-Mito database as a protein-protein interaction (PPI) prediction data archive, including prediction results for exhaustive protein pairs of 654 mitochondria-related human proteins. MEGADOCK-Web-Mito enables users to search for all PPI prediction results efficiently and comprehensively. In par…
Mount Sinai School of Medicine（西奈山伊坎医学院）；University of Minnesota（明尼苏达大学）
Protein interaction networks are a promising type of data for studying complex biological systems. However, despite the rich information embedded in these networks, they face important data quality challenges of noise and incompleteness that adversely affect the results obtained from their analysis. Here, we explore the use of the concept of common neighborhood similarity (CNS), which is a form of local structure in networks, to address these issues. Although several CNS measures have been proposed in the literature, an understanding of their relative efficacies for the analysis of interaction…
