Figure 1.

GPU-accelerated peptide scoring algorithm. (A) Differences in architecture differences between CPU (left) and GPU (right). Traditionally, computations are performed by transferring data from global memory to the CPU cores. GPU computing allows for the transfer of data from global memory to device memory, where the GPU cores can access them. (B) Pseudocode of the two main GPU parallel functions. Algorithm 1 (top) describes the look-up table generation of the peptides and algorithm 2 (bottom) describes the calculation of the agonist likelihood scores.