High-throughput Bayesian network learning using heterogeneous multicore computers

摘要: Aberrant intracellular signaling plays an important role in many diseases. The causal structure of signal transduction networks can be modeled as Bayesian Networks (BNs), and computationally learned from experimental data. However, learning the (BNs) is NP-hard problem that, even with fast heuristics, too time consuming for large, clinically (20--50 nodes). In this paper, we present a novel graphics processing unit (GPU)-accelerated implementation Monte Carlo Markov Chain-based algorithm BNs that up to 7.5-fold faster than current general-purpose processor (GPP)-based implementations.The GPU-based just one several implementations within larger application, each optimized different input or machine configuration. We describe methodology use build extensible assembled these variants, target broad range heterogeneous systems, e.g., GPUs, multicore GPPs. Specifically show how Merge programming model efficiently integrate, test intelligently select among potential implementations.