Parallelizing Peptide-Spectrum scoring using modern graphics processing units

Tandem mass spectrometry is a powerful experimental tool used in molecular biology to determine the composition of protein mixtures. In a tandem mass experiment, peptide ion selection algorithms generally select only the most abundant peptide ions for further fragmentation. Because of this, the low-abundance proteins in a sample rarely get identified. A Real-Time Peptide-Spectrum Matching algorithm (RT-PSM) was introduced to achieve real-time peptide identification for solving this abundance related biases. Profiling results show that the Peptide-Spectrum similarity scoring is one of the most time-consuming module of RT-PSM. In this study, we develop a parallel algorithm for Peptide-Spectrum scoring using NVIDIA CUDA technology. As RT-PSM employs a scoring function based on shared peak counts, our algorithm can also be applied to other software that uses similar scoring schemes. Moreover, we introduce an algorithm to reduce the number of comparisons in calculating shared peak counts. In addition, as the CUDA architecture is unique, we introduce optimizations for the CUDA architecture to achieve better performance. A simulation shows a 190-fold speedup on the scoring module and a 26-fold speedup on the entire process. The developed algorithm can be employed to develop real-time control methods for tandem mass spectrometry.