Parallel branch-and-bound approach with MPI technology in inferring chemical compounds with path frequency

摘要: Drug design is the approach of finding drugs by using computational tools. When designing a new drug, structure drug molecule can be modeled classification potential chemical compounds. Kernel Methods have been successfully used in classifying Frequency labeled paths has proposed to map compounds into feature order classify characteristics target In this study, we an algorithm based on method via parallel computing technology reduce computation time. This less constrain timing allows us aim at back tracking full scheme all possible pre-images, regardless their difference molecular structure, only if they shared with same vector. Our modified BB-CIPF and MPI The experimental results show that our algorithms time effectively for compound inference problem.