MOTIVATION: Increasingly comprehensive characterization of cancer-associated genetic alterations has paved the way for the development of highly specific therapeutic vaccines. Predicting precisely the binding and presentation of peptides to major histocompatibility complex (MHC) alleles is an important step toward such therapies. Recent data suggest that presentation of both class I and II epitopes are critical for the induction of a sustained effective immune response. However, the prediction performance for MHC class II has been limited compared to class I. RESULTS: We present a transformer neural network model which leverages self-supervised pretraining from a large corpus of protein sequences. We also propose a multiple instance learning (MIL) framework to deconvolve mass spectrometry data where multiple potential MHC alleles may have presented each peptide. We show that pretraining boosted the performance for these tasks. Combining pretraining and the novel MIL approach, our model outperforms state-of-the-art models based on peptide and MHC sequence only for both binding and cell surface presentation predictions. AVAILABILITY AND IMPLEMENTATION: Our source code is available at https://github.com/s6juncheng/BERTMHC under a noncommercial license. A webserver is available at https://bertmhc.privacy.nlehd.de/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Various algorithms have been proposed for finding a Bayesian network structure that is guaranteed to maximize a given scoring function. Implementations of state-of-the-art algorithms, solvers , for this Bayesian network structure learning problem rely on adaptive search strategies, such as branch-and-bound and integer linear programming techniques. Thus, the time requirements of the solvers are not well characterized by simple functions of the instance size. Furthermore, no single solver dominates the others in speed. Given a problem instance, it is thus a priori unclear which solver will perform best and how fast it will solve the instance. We show that for a given solver the hardness of a problem instance can be efficiently predicted based on a collection of non-trivial features which go beyond the basic parameters of instance size. Specifically, we train and test statistical models on empirical data, based on the largest evaluation of state-of-the-art exact solvers to date. We demonstrate that we can predict the runtimes to a reasonable degree of accuracy. These predictions enable effective selection of solvers that perform well in terms of runtimes on a particular instance. Thus, this work contributes a highly efficient portfolio solver that makes use of several individual solvers.

A recent breadth-first branch and bound algorithm (BFBnB)for learning Bayesian network structures (Maloneet al. 2011) uses two bounds to prune the searchspace for better efficiency; one is a lower bound calculatedfrom pattern database heuristics, and the otheris an upper bound obtained by a hill climbing search.Whenever the lower bound of a search path exceeds theupper bound, the path is guaranteed to lead to suboptimalsolutions and is discarded immediately. This paperintroduces methods for tightening the bounds. Thelower bound is tightened by using more informed variablegroupings when creating the pattern databases, andthe upper bound is tightened using an anytime learningalgorithm. Empirical results show that these boundsimprove the efficiency of Bayesian network learning bytwo to three orders of magnitude.

There are various algorithms for finding a Bayesian networkstructure (BNS) that is optimal with respect to a given scoring function. No single algorithm dominates the others in speed, and, given a problem instance, it is a priori unclear which algorithm will perform best and how fast it will solve the problem. Estimating the runtimes directly is extremely difficult as they are complicated functions of the instance. The main contribution of this paper is characterization of the empirical hardness of an instance for a given algorithm based on a novel collection of non-trivial, yet efficiently computable features. Our empirical results, based on the largest evaluation of state-of-the-art BNS learning algorithms to date, demonstrate that we can predict the runtimes to a reasonable degree of accuracy, and effectively select algorithms that perform well on a particular instance. Moreover, we also show how the results can be utilized in building a portfolio algorithm that combines several individual algorithms in an almost optimal manner.

Bayesian network structure learning is the well-known computationally hard problem of finding a directed acyclic graph structure that optimally describes given data. A learned structure can then be used for probabilistic inference. While exact inference in Bayesian networks is in general NP-hard, it is tractable in networks with low treewidth. This provides good motivations for developing algorithms for the NP-hard problem of learning optimal bounded treewidth Bayesian networks (BTW-BNSL). In this work, we develop a novel score-based approach to BTW-BNSL, based on casting BTW-BNSL as weighted partial Maximum satisfiability. We demonstrate empirically that the approach scales notably better than a recent exact dynamic programming algorithm for BTW-BNSL.

In this paper, an artificial neural network (ANN) model is proposed to predict the flexibility (or robustness against system load fluctuations in heterogeneous computing systems) of dynamic loop scheduling (DLS) methods. The multilayer perceptron (MLP) ANN model has been used to predict the degree of robustness of a DLS method, given specific values for the problem size, the system size, and the characteristics of the system load fluctuations as a compound effect of the variations in the application's iteration execution times and the processor availabilities. The developed MLP ANN model can be useful in an effective selection of the most robust DLS technique for scheduling a certain type of scientific application onto a given set of non-dedicated heterogeneous processors, when their system load is expected to fluctuate unpredictably during the application's runtime.

In this paper, learning a Bayesian network structure that optimizes a scoring function for a given dataset is viewed as a shortest path problem in an implicit state-space search graph. This perspective highlights the importance of two research issues: the development of search strategies for solving the shortest path problem, and the design of heuristic functions for guiding the search. This paper introduces several techniques for addressing the issues. One is an A* search algorithm that learns an optimal Bayesian network structure by only searching the most promising part of the solution space. The others are mainly two heuristic functions. The first heuristic function represents a simple relaxation of the acyclicity constraint of a Bayesian network. Although admissible and consistent, the heuristic may introduce too much relaxation and result in a loose bound. The second heuristic function reduces the amount of relaxation by avoiding directed cycles within some groups of variables. Empirical results show that these methods constitute a promising approach to learning optimal Bayesian network structures.

In this work, we empirically evaluate the capability of various scoring functions of Bayesian networks for recovering true underlying structures. Similar investigations have been carried out before, but they typically relied on approximate learning algorithms to learn the network structures. The suboptimal structures found by the approximation methods have unknown quality and may affect the reliability of their conclusions. Our study uses an optimal algorithm to learn Bayesian network structures from datasets generated from a set of gold standard Bayesian networks. Because all optimal algorithms always learn equivalent networks, this ensures that only the choice of scoring function affects the learned networks. Another shortcoming of the previous studies stems from their use of random synthetic networks as test cases. There is no guarantee that these networks reflect real-world data. We use real-world data to generate our gold-standard structures, so our experimental design more closely approximates real-world situations. A major finding of our study suggests that, in contrast to results reported by several prior works, the Minimum Description Length (MDL) (or equivalently, Bayesian information criterion (BIC)) consistently outperforms other scoring functions such as Akaike's information criterion (AIC), Bayesian Dirichlet equivalence score (BDeu), and factorized normalized maximum likelihood (fNML) in recovering the underlying Bayesian network structures. We believe this finding is a result of using both datasets generated from real-world applications rather than from random processes used in previous studies and learning algorithms to select high-scoring structures rather than selecting random models. Other findings of our study support existing work, e.g., large sample sizes result in learning structures closer to the true underlying structure; the BDeu score is sensitive to the parameter settings; and the fNML performs pretty well on small datasets. We also tested a greedy hill climbing algorithm and observed similar results as the optimal algorithm.

We describe a memory-efficient implementation of a dynamic programming algorithm for learning the optimal structure of a Bayesian network from training data. The algorithm leverages the layered structure of the dynamic programming graphs representing the recursive decomposition of the problem to reduce the memory requirements of the algorithm from O(n2n) to O(C(n, n/2)), where C(n, n/2) is the binomial coefficient. Experimental results show that the approach runs up to an order of magnitude faster and scales to datasets with more variables than previous approaches.

BACKGROUND: High-throughput mass spectrometry (MS) proteomics data is increasingly being used to complement traditional structural genome annotation methods. To keep pace with the high speed of experimental data generation and to aid in structural genome annotation, experimentally observed peptides need to be mapped back to their source genome location quickly and exactly. Previously, the tools to do this have been limited to custom scripts designed by individual research groups to analyze their own data, are generally not widely available, and do not scale well with large eukaryotic genomes. RESULTS: The Proteogenomic Mapping Tool includes a Java implementation of the Aho-Corasick string searching algorithm which takes as input standardized file types and rapidly searches experimentally observed peptides against a given genome translated in all 6 reading frames for exact matches. The Java implementation allows the application to scale well with larger eukaryotic genomes while providing cross-platform functionality. CONCLUSIONS: The Proteogenomic Mapping Tool provides a standalone application for mapping peptides back to their source genome on a number of operating system platforms with standard desktop computer hardware and executes very rapidly for a variety of datasets. Allowing the selection of different genetic codes for different organisms allows researchers to easily customize the tool to their own research interests and is recommended for anyone working to structurally annotate genomes using MS derived proteomics data.

BACKGROUND: This paper presents a framework for integrating disparate data sets to predict gene function. The algorithm constructs a graph, called an integrated similarity graph, by computing similarities based upon both gene expression and textual phenotype data. This integrated graph is then used to make predictions about whether individual genes should be assigned a particular annotation from the Gene Ontology. RESULTS: A combined graph was generated from publicly-available gene expression data and phenotypic information from Saccharomyces cerevisiae. This graph was used to assign annotations to genes, as were graphs constructed from gene expression data and textual phenotype information alone. While the F-measure appeared similar for all three methods, annotations based upon the integrated similarity graph exhibited a better overall precision than gene expression or phenotype information alone can generate. The integrated approach was also able to assign almost as many annotations as the gene expression method alone, and generated significantly more total and correct assignments than the phenotype information could provide. CONCLUSION: These results suggest that augmenting standard gene expression data sets with publicly-available textual phenotype data can help generate more precise functional annotation predictions while mitigating the weaknesses of a standard textual phenotype approach.