Cell-surface proteins play a critical role in cell function and are primary targets for therapeutics. CITE-seq is a single-cell technique that enables simultaneous measurement of gene and surface protein expression. It is powerful but costly and technically challenging. Computational methods have been developed to predict surface protein expression using gene expression information such as from single-cell RNA sequencing (scRNA-seq) data. Existing methods however are computationally demanding and lack the interpretability to reveal underlying biological processes. We propose CrossmodalNet, an interpretable machine learning model, to predict surface protein expression from scRNA-seq data. Our model with a customized adaptive loss accurately predicts surface protein abundances. When samples from multiple time points are given, our model encodes temporal information into an easy-to-interpret time embedding to make prediction in a time-point-specific manner, and is able to uncover noise-free causal gene-protein relationships. Using three publicly available time-resolved CITE-seq data sets, we validate the performance of our model by comparing it with benchmarking methods and evaluate its interpretability. Together, we show that our method accurately and interpretably profiles surface protein expression using scRNA-seq data, thereby expanding the capacity of CITE-seq experiments for investigating molecular mechanisms involving surface proteins.

BACKGROUND: Mitochondria play essential roles in regulating cellular functions. Some drug treatments and molecular interventions have been reported to have off-target effects damaging mitochondria and causing severe side effects. The development of a database for the management of mitochondrial toxicity-related molecules and their targets is important for further analyses. RESULTS: To correlate chemical, biological and mechanistic information on clinically relevant mitochondria-related toxicity, a comprehensive mitochondrial toxicity database (MitoTox) was developed. MitoTox is an electronic repository that integrates comprehensive information about mitochondria-related toxins and their targets. Information and data related to mitochondrial toxicity originate from various sources, including scientific journals and other electronic databases. These resources were manually verified and extracted into MitoTox. The database currently contains over 1400 small-molecule compounds, 870 mitochondrial targets, and more than 4100 mitochondrial toxin-target associations. Each MitoTox data record contains over 30 fields, including biochemical properties, therapeutic classification, target proteins, toxicological data, mechanistic information, clinical side effects, and references. CONCLUSIONS: MitoTox provides a fully searchable database with links to references and other databases. Potential applications of MitoTox include toxicity classification, prediction, reference and education. MitoTox is available online at http://www.mitotox.org .

Summary Multi‐core systems equipped with micro processing units and accelerators such as digital signal processors (DSPs) and graphics processing units (GPUs) have become a major trend in processor design in recent years in attempts to meet ever‐increasing application performance requirements. Open Computing Language (OpenCL) is one of the programming languages that include new extensions proposed to exploit the computing power of these kinds of processors. Among the newly extended language features, the single‐instruction multiple‐data (SIMD) linguistics and vector types are added to OpenCL to exploit hardware features of the accelerators. The addition makes it necessary to consider how traditional compiler data flow analysis can be adopted to meet the optimization requirements of vector linguistics. In this paper, we propose a calculus framework to support the data flow analysis of vector constructs for OpenCL programs that compilers can use to perform SIMD optimizations. We model OpenCL vector operations as data access functions in the style of mathematical functions. We then show that the data flow analysis for OpenCL vector linguistics can be performed based on the data access functions. Based on the information gathered from data flow analysis, we illustrate a set of SIMD optimizations on OpenCL programs. The experimental results incorporating our calculus and our proposed compiler optimizations show that the proposed SIMD optimizations can provide average performance improvements of 22% on x86 CPUs and 4% on advanced micro devices GPUs. For the selected 15 benchmarks, 11 of them are improved on x86 CPUs, and six of them are improved on advanced micro devices GPUs. The proposed framework has the potential to be used to construct other SIMD optimizations on OpenCL programs. Copyright © 2015 John Wiley & Sons, Ltd.

As microprocessors evolve into heterogeneous architectures with multi-cores of MPUs and GPUs, programming model supports become important for programming such architectures. To address this issue, OpenCL is proposed. Currently, most of OpenCL implementations take LLVM as their infrastructures. This presents an opportunity to demonstrate whether OpenCL can be effectively implemented on other compiler infrastructures. For example, Open64, which is another open source compiler and known to generate efficient codes for microprocessors, can contribute further to performance improvements and enhancing the adoption of heterogeneous computing based on OpenCL. In this paper, we describe the flow to enable an OpenCL compiler based on Open64 infrastructures for ATI GPUs. Our work includes the extension of the front-end parser for OpenCL, the generation of high-level intermediate representations with OpenCL linguistics, performing high-level optimization, and finally applying OpenCL specific optimization for code generations. Preliminary experimental results show that our compiler based on Open64 is able to generate efficient codes for OpenCL programs.

Probabilistic roadmap planners (PRMs) are widely used in high dimensional motion planning problems. However, they are less effective in solving narrow passages because feasible configurations in a thin space can rarely be sampled by random. Although some approaches have been proposed, they are either involved in complicated geometrical computations or requiring much information of obstacles. Moreover, if the configuration spaces are dynamic instead of fixed, some solutions may be failure in some unexpected situations. In this work, we provide a novel approach to replace the randomized sampler with the Gaussian PRM sampler. For dynamic configuration spaces, we also invite a machine learning technique to dynamically classify the pre-built samplers for different spaces

Collision detection is a critical module in many applications such as computer graphics, robot motion planning, physical simulation, CAD/CAM, and molecular modeling. Many efficient algorithms have been proposed to solve the collision-detection problem, and most of them use some sort of hierarchical bounding volume to speed up the time-consuming process. In this research, we focus on a special situation for applications such as interactive virtual environment, where accuracy may not be crucial but the available time for performing the check is limited. We describe our findings on the inherencies of bounding volume traverse trees (BVTT). Based on this observation, we propose a time-budgeted collision detection method that can traverse the critical regions of BVTT earlier if the time budget is limited. Preliminary experimental results are also reported, and a new strategy is suggested based on the concept of confidence value