MOTIVATION: The goal of oligonucleotide (oligo) design is to select oligos that optimize a set of design criteria. Oligo design problems are combinatorial in nature and require computationally intensive models to evaluate design criteria. Even relatively small problems can be intractable for brute-force approaches that test every possible combination of oligos, so heuristic approaches must be used to find near-optimal solutions. RESULTS: We present a general reinforcement learning (RL) framework, called OligoRL, to solve oligo design problems with complex constraints. OligoRL allows 'black-box' design criteria and can be adapted to solve many oligo design problems. We highlight the flexibility of OligoRL by building tools to solve three distinct design problems: (i) finding pools of random DNA barcodes that lack restriction enzyme recognition sequences (CutFreeRL); (ii) compressing large, non-degenerate oligo pools into smaller degenerate ones (OligoCompressor) and (iii) finding Not-So-Random hexamer primer pools that avoid rRNA and other unwanted transcripts during RNA-seq library preparation (NSR-RL). OligoRL demonstrates how RL offers a general solution for complex oligo design problems. AVAILABILITY AND IMPLEMENTATION: OligoRL and all simulation codes are available as a Julia package at http://jensenlab.net/tools and archived at https://archive.softwareheritage.org/browse/origin/directory/?origin_url=https://github.com/bmdavid2/OligoRL. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Gapsplit generates random samples from convex and non-convex constraint-based models by targeting under-sampled regions of the solution space. Gapsplit provides uniform coverage of linear, mixed-integer and general non-linear models. AVAILABILITY AND IMPLEMENTATION: Python and Matlab source code are freely available at http://jensenlab.net/tools. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Abstract Motivation: Metabolic reaction maps allow visualization of genome-scale models and high-throughput data in a format familiar to many biologists. However, creating a map of a large metabolic model is a difficult and time-consuming process. MetDraw fully automates the map-drawing process for metabolic models containing hundreds to thousands of reactions. MetDraw can also overlay high-throughput ‘omics’ data directly on the generated maps. Availability and implementation: Web interface and source code are freely available at http://www.metdraw.com. Contact: [email protected] Supplementary information: Supplementary data are available at Bioinformatics online.

MOTIVATION: Flux balance analysis (FBA) has been used extensively to analyze genome-scale, constraint-based models of metabolism in a variety of organisms. The predictive accuracy of such models has recently been improved through the integration of high-throughput expression profiles of metabolic genes and proteins. However, extensions of FBA often require that such data be discretized a priori into sets of genes or proteins that are either 'on' or 'off'. This procedure requires selecting relatively subjective expression thresholds, often requiring several iterations and refinements to capture the expression dynamics and retain model functionality. RESULTS: We present a method for mapping expression data from a set of environmental, genetic or temporal conditions onto a metabolic network model without the need for arbitrary expression thresholds. Metabolic Adjustment by Differential Expression (MADE) uses the statistical significance of changes in gene or protein expression to create a functional metabolic model that most accurately recapitulates the expression dynamics. MADE was used to generate a series of models that reflect the metabolic adjustments seen in the transition from fermentative- to glycerol-based respiration in Saccharomyces cerevisiae. The calculated gene states match 98.7% of possible changes in expression, and the resulting models capture functional characteristics of the metabolic shift. AVAILABILITY: MADE is implemented in Matlab and requires a mixed-integer linear program solver. Source code is freely available at http://www.bme.virginia.edu/csbl/downloads/.

Coordinating distributed executions is achieved by two widely used approaches: process groups and transactions. Typically, the two represent a trade-off in terms of the degrees of consistency and performance. By applying transaction concurrency control techniques to characterize and design process group multicast orderings, we aim to provide aspects of both ends of the trade-off. In particular, we propose a framework in which each message multicast is regarded as a transaction. Appropriate message ordering protocols are devised and shown to be correct using a variant of concurrency control theory. Also, we are able do incorporate certain aspects of application semantics for which existing process group approaches are inadequate. Finally, our framework provides a means to characterize the performance of orderings to allow a comparison of different ordering protocols.

Abstract An extensive literature survey reveals that computational analyses of pure minimumcost‐network algorithms do not address the use of dual incremental codes. Previous studies only compare dual‐arc‐infeasible‐, primal‐dual‐, and primal‐type network codes. Of these three types, the primal method has been shown to be computationally superior to the other methods. This paper surveys the historical development of the dual incremental method up to and including the code DUALINC. Following a comprehensive description of DUALINC, a computational comparison is made with three other recently publicized codes, two primal codes and a dual‐arc‐infeasible code. The results of this comparison reveal that the new dual incremental code is much more efficient than the dual‐arc‐infeasible code and that it is competitive with the primal codes.