Java bytecode decompilation is a process of reverse translation that restores Java source code by the corresponding bytecode. Java bytecode is an intermediate representation based on abstract stack machine. It may have arbitrary control flow graph, whereas the Java language contains control structures that always form a strict hierarchy. Decompilation aims at restoring all control structures, including Java exception handling blocks. In the Excelsior RVM (Java Virtual Machine with a static compiler), bytecode is decompiled in a structural intermediate representation for further optimization. When building exception handling blocks, the Excelsior RVM’s compiler assumes that the bytecode must be emitted by the standard Java source to bytecode compiler and uses a few heuristics to make reverse transformation. However, it is not always possible if the bytecode is produced by other instruments. This paper presents a decompilation algorithm that produces exception handling blocks given any correct bytecode. The algorithm has been implemented, integrated into the Excelsior RVM and tested on real-world applications.

This paper presents some popular classical methods used for time series analysis and prediction. At first relatively simple models of averaging and smoothing are described, then autoregressive and moving average models, and a “mixed” autoregressive-moving-average model as a result of crossing of the latter two mentioned models. At last an autoregressive integrated moving average model is described.

Before RNA transcription starts, special segments of DNA form a complex of regulatory proteins called transcription factors. This complex allows RNA polymerase to be bound to DNA and to start reading RNA. It is a difficult problem to search binding sites on DNA because of many factors influencing the binding. In particular, other sites in the vicinity of a given site may influence binding. To reveal those dependencies the authors introduce histograms of density distribution of binding sites, called genomic profiles. The software package developed in the scope of this work allows building genomic profiles using the prediction of binding sites by a weight matrices algorithm for different implementations: for multicore CPU’s or NVidia GPU’s using CUDA. In addition, the software allows clustering genomic profiles using K-means clustering and hierarchical clustering. The algorithm allows to build samples – random transcription factors hierarchies based on the existing experimental structural classification to estimate the correspondence between genomic profiles construction and the existing classification. The analysis of correspondence of genomic profiles with biological classification of transcription factors was developed using the sampling algorithm.

The contribution of the paper is to clarify connections between real-time models of concurrency. In particular, we defined a category of timed causal trees and investigated how it relates to other categories of timed models. Moreover, using a larger model called timed event trees, we constructed an adjunction from the category of timed causal trees to the category of timed event structures. Thereby we showed that timed causal trees are more trivial than timed event structures because they reflect only one aspect of true concurrency, causality, and they apply causality without a notion of event. On the other hand, the first model is more expressive than the latter in that possible runs of a timed causal tree can be defined in terms of a tree without restrictions, but the set of the possible runs of any event structure must be closed under the shuffling of concurrent transitions.

This paper describes how the SPIN model checker can be applied to solving puzzles, such as riverIQGame (an advanced “wolf, goat and cabbage” puzzle) or “Irregular IQ Cube” (also known as ‘Square-1’).