Mobile-edge computing (MEC) brings the potential to address the ever increasing computation demands from the mobile users (MUs). In addition to local processing, the resource-constrained MUs in an MEC system can also offload computation to the nearby servers for remote execution. With the explosive growth of mobile devices, computation offloading faces the challenge of spectrum congestion, which, in turn, deteriorates the overall quality of computation experience. This article, hence, investigates computation task scheduling in a heterogeneous cellular and WiFi MEC system. Such a system provides both licensed and unlicensed spectrum opportunities. Due to the sharing of communication and computation resources as well as the uncertainties, we formulate the problem of computation task scheduling among the competing MUs in a stationary heterogeneous edge computing system as a noncooperative stochastic game. We propose an approximation-based multiagent Markov decision process without the global system state observations, under which a multiagent proximal policy optimization (PPO) algorithm is derived to solve the corresponding Nash equilibrium. When expanding to a nonstationary heterogeneous edge computing system, the obtained algorithm suffers from the slow convergence due to constrained adaptability. Accordingly, we explore meta-learning and propose a multiagent meta-PPO algorithm, which rapidly adapts the control policy learning to the nonstationarity. Numerical experiments demonstrate performance gains from our proposed algorithms.

Mobile edge computing (MEC) is a promising paradigm for alleviating the computation burden of resource-constrained mobile devices. Nevertheless, the majority of existing efforts concentrate on offloading computations from mobile de-vices to an edge computing server through the cellular networks only. With the development of wireless connectivity technologies, WiFi networks over unlicensed spectrum provide a “green” (i.e., cost-efficient and economical) alternative for computation offloading. In this paper, we investigate the problem of computation offloading in a heterogeneous WiFi and cellular MEC system, where both the WiFi and the cellular networks are possible for offloading the arriving computation tasks at a mobile user (MU). The objective of an MU is to minimize the long-term cost, which can be described as a single-agent Markov decision process (MDP) by accounting for the inherent system dynamics in the MU mobility, sporadic computation task arrivals and wireless connectivity variations. To solve the optimal strategy for the formulated MDP with a high-dimensional state space but without the statistical knowledge of system dynamics, we resort to a model-free deep reinforcement learning algorithm. Numerical experiments verify that the proposed algorithm is able to significantly reduce the average computation offloading cost compared with other baselines.

MOTIVATION: A large number of proteins contain metal ions that are essential for their stability and biological activity. Identifying and characterizing metal-binding sites through computational methods is necessary when experimental clues are lacking. Almost all published computational methods are designed to distinguish metal-binding sites from non-metal-binding sites. However, discrimination between different types of metal-binding sites is also needed to make more accurate predictions. RESULTS: In this work, we proposed a novel algorithm called mFASD, which could discriminate different types of metal-binding sites effectively based on 3D structure data and is useful for accurate metal-binding site prediction. mFASD captures the characteristics of a metal-binding site by investigating the local chemical environment of a set of functional atoms that are considered to be in contact with the bound metal. Then a distance measure defined on functional atom sets enables the comparison between different metal-binding sites. The algorithm could discriminate most types of metal-binding sites from each other with high sensitivity and accuracy. We showed that cascading our method with existing ones could achieve a substantial improvement of the accuracy for metal-binding site prediction. AVAILABILITY AND IMPLEMENTATION: Source code and data used are freely available from http://staff.ustc.edu.cn/∼liangzhi/mfasd/

MOTIVATION: Zinc-binding proteins are the most abundant metallo-proteins in Protein Data Bank (PDB). Accurate prediction of zinc-binding sites in proteins of unknown function may provide important clues for the inference of protein function. As zinc binding is often associated with characteristic 3D arrangements of zinc ligand residues, its prediction may benefit from using not only the sequence information but also the structure information of proteins. RESULTS: In this work, we present a structure-based method, TEMSP (3D TEmplate-based Metal Site Prediction), to predict zinc-binding sites. TEMSP significantly improves over previously reported best methods in predicting as many as possible true ligand residues for zinc with minimum overpredictions: if only those results in which all zinc ligand residues have been correctly predicted are defined as true positives, our method improves sensitivity from less than 30% to above 60%, and selectivity from around 25% to 80%. These results are for predictions based on apo state structures. In addition, the method can predict the zinc-bound local structures reliably, generating predictions useful for function inference. We applied TEMSP to 1888 protein structures of the 'Unknown Function' class in the PDB database. A number of zinc-binding sites have been discovered de novo, i.e. based solely on the protein structures. Using the predicted local structures of these sites, possible functional roles were analyzed. AVAILABILITY: TEMSP is freely available from http://netalign.ustc.edu.cn/temsp/.

UNLABELLED: NetAlign is a web-based tool designed to enable comparative analysis of protein interaction networks (PINs). NetAlign compares a query PIN with a target PIN by combining interaction topology and sequence similarity to identify conserved network substructures (CoNSs), which may derive from a common ancestor and disclose conserved topological organization of interactions in evolution. To exemplify the application of NetAlign, we perform two genome-scale comparisons with (1) the Escherichia coli PIN against the Helicobacter pylori PIN and (2) the Saccharomyces cerevisiae PIN against the Caenorrhabditis elegans PIN. Many of the identified CoNSs correspond to known complexes; therefore, cross-species PIN comparison provides a way for discovery of conserved modules. In addition, based on the species-to-species differences in CoNSs, we reformulate the problems of protein-protein interaction (PPI) prediction and species divergence from a network perspective. AVAILABILITY: http://www1.ustc.edu.cn/lab/pcrystal/NetAlign.

BACKGROUND: Recent progresses in high-throughput proteomics have provided us with a first chance to characterize protein interaction networks (PINs), but also raised new challenges in interpreting the accumulating data. RESULTS: Motivated by the need of analyzing and interpreting the fast-growing data in the field of proteomics, we propose a comparative strategy to carry out global analysis of PINs. We compare two PINs by combining interaction topology and sequence similarity to identify conserved network substructures (CoNSs). Using this approach we perform twenty-one pairwise comparisons among the seven recently available PINs of E.coli, H.pylori, S.cerevisiae, C.elegans, D.melanogaster, M.musculus and H.sapiens. In spite of the incompleteness of data, PIN comparison discloses species conservation at the network level and the identified CoNSs are also functionally conserved and involve in basic cellular functions. We investigate the yeast CoNSs and find that many of them correspond to known complexes. We also find that different species harbor many conserved interaction regions that are topologically identical and these regions can constitute larger interaction regions that are topologically different but similar in framework. Based on the species-to-species difference in CoNSs, we infer potential species divergence. It seems that different species organize orthologs in similar but not necessarily the same topology to achieve similar or the same function. This attributes much to duplication and divergence of genes and their associated interactions. Finally, as the application of CoNSs, we predict 101 protein-protein interactions (PPIs), annotate 339 new protein functions and deduce 170 pairs of orthologs. CONCLUSION: Our result demonstrates that the cross-species comparison strategy we adopt is powerful for the exploration of biological problems from the perspective of networks.