In order to directly observe the refolding kinetics from a partially misfolded state to a native state in the bottom of the protein-folding funnel, we used a “caging” strategy to trap the β-sheet structure of ubiquitin in a misfolded conformation. We used molecular dynamics simulation to generate the cage-induced, misfolded structure and compared the structure of the misfolded ubiquitin with native ubiquitin. Using laser flash irradiation, the cage can be cleaved from the misfolded structure within one nanosecond, and we monitored the refolding kinetics of ubiquitin from this misfolded state to the native state by photoacoustic calorimetry and photothermal beam deflection techniques on nanosecond to millisecond timescales. Our results showed two refolding events in this refolding process. The fast event is shorter than 20 ns and corresponds to the instant collapse of ubiquitin upon cage release initiated by laser irradiation. The slow event is ~60 μs, derived from a structural rearrangement in β-sheet refolding. The event lasts 10 times longer than the timescale of β-hairpin formation for short peptides as monitored by temperature jump, suggesting that rearrangement of a β-sheet structure from a misfolded state to its native state requires more time than ab initio folding of a β-sheet.

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. Electronic supplementary materialThe online version of this article (doi:10.1186/s13321-017-0238-7) contains supplementary material, which is available to authorized users.