G.A.M.E.: GPU-accelerated mixture elucidator.

GPU acceleration is useful in solving complex chemical information problems. Identifying unknown structures from the mass spectra of natural product mixtures has been a desirable yet unresolved issue in metabolomics. However, this elucidation process has been hampered by complex experimental data and the inability of instruments to completely separate different compounds. Fortunately, with current high-resolution mass spectrometry, one feasible strategy is to define this problem as extending a scaffold database with sidechains of different probabilities to match the high-resolution mass obtained from a high-resolution mass spectrum. By introducing a dynamic programming (DP) algorithm, it is possible to solve this NP-complete problem in pseudo-polynomial time. However, the running time of the DP algorithm grows by orders of magnitude as the number of mass decimal digits increases, thus limiting the boost in structural prediction capabilities. By harnessing the heavily parallel architecture of modern GPUs, we designed a "compute unified device architecture" (CUDA)-based GPU-accelerated mixture elucidator (G.A.M.E.) that considerably improves the performance of the DP, allowing up to five decimal digits for input mass data. As exemplified by four testing datasets with verified constitutions from natural products, G.A.M.E. allows for efficient and automatic structural elucidation of unknown mixtures for practical procedures. Graphical abstract .

An efficient computer-aided structural elucidation strategy for mixtures using an iterative dynamic programming algorithm.

The identification of chemical structures in natural product mixtures is an important task in drug discovery but is still a challenging problem, as structural elucidation is a time-consuming process and is limited by the available mass spectra of known natural products. Computer-aided structure elucidation (CASE) strategies seek to automatically propose a list of possible chemical structures in mixtures by utilizing chromatographic and spectroscopic methods. However, current CASE tools still cannot automatically solve structures for experienced natural product chemists. Here, we formulated the structural elucidation of natural products in a mixture as a computational problem by extending a list of scaffolds using a weighted side chain list after analyzing a collection of 243,130 natural products and designed an efficient algorithm to precisely identify the chemical structures. The complexity of such a problem is NP-complete. A dynamic programming (DP) algorithm can solve this NP-complete problem in pseudo-polynomial time after converting floating point molecular weights into integers. However, the running time of the DP algorithm degrades exponentially as the precision of the mass spectrometry experiment grows. To ideally solve in polynomial time, we proposed a novel iterative DP algorithm that can quickly recognize the chemical structures of natural products. By utilizing this algorithm to elucidate the structures of four natural products that were experimentally and structurally determined, the algorithm can search the exact solutions, and the time performance was shown to be in polynomial time for average cases. The proposed method improved the speed of the structural elucidation of natural products and helped broaden the spectrum of available compounds that could be applied as new drug candidates. A web service built for structural elucidation studies is freely accessible via the following link ( http://csccp.cmdm.tw/ ).