The interprotein scoring noises in glide docking scores

Abstract

Small molecule drugs are rarely selective enough to interact solely with their designated targets. Unintended "off-target" interactions often lead to side effects, but also serendipitously lead to new therapeutic uses. Identification of the off-targets of a compound is therefore of significant value to the evaluation of its developmental potential. In computational biology, the strategy of "reverse docking" has been introduced to predict the targets of a compound, which uses a compound to virtually screen a library of proteins, reversing the bait and prey in "normal" docking screenings. The present study shows that, in reverse docking, additional optimization of the scoring function may help to improve the target prediction accuracy. In a case study with the Glide scores, we found that only 57% of the ligand-protein relationships could be correctly identified in a library of 58 complexes whose crystal binding conformations were all able to be accurately reproduced. This was likely a result of the constant over- or under-estimation of the scores for specific proteins. In other words, there were interprotein noises in the Glide scores. Introducing a correction term based on protein characteristics improved the target-prediction accuracy by 27% (57-72%). It is our hope that this focused discussion on the Glide scores would invite further efforts to characterize and normalize this type of interprotein noises in all docking scores, so that better target prediction accuracy can be achieved with the strategy of reverse docking.

Improvements, trends, and new ideas in molecular docking: 2012–2013 in review

“…By testing scoring and ranking powers separately, the authors found that the good performance in one did not necessarily deliver a good performance in another. This finding relates to other observations of task-specific performance of many scoring functions (e.g., for GlideScore [71] and HYdrogen bond and DEhydration energies (HYDE [72] )) and points to the need to tailor their use, following appropriate evaluation for specific targets or tasks.…”

“…Many studies have now demonstrated that using an ensemble approach is superior to a single receptor conformation input. [16] However, the main (71) SR c = 51-69% [212] Rank Custom set (6) ROC AUC = 0.76…”

Improving inverse docking target identification with Z‐score selection

“…Existing docking algorithm scoring functions have been optimized to accomplish its primary purpose: ranking the true positive ligands toward the top of the list of sorted docking scores within one target receptor. However, limitations of these molecular docking methods emerged when they were applied to inverse docking, that is, when the same set of ligands was docked into multiple target receptors (Luo et al, 2017;Schomburg et al, 2014;Wang, Zhou, et al, 2012). To investigate optimal ways of selecting true positive ligand-protein pairs from the inverse docking results, we worked with three different databases: Astex, DUD, and DUD-E. Our Astex dataset was comprised of 85 unique ligand-protein pairs, from the DUD we collected 8 different proteins each paired with 11-74 active ligands, and from the DUD-E we selected 102 diverse proteins each paired with 20 unique active ligands.…”

Prediction of Protein-Protein Docking Sites Based on a Cloud-Computing Pipeline

“…The study of protein-protein docking sites not only reveals the relationship between protein and protein, but also contributes to protein engineering such as molecular design and computer-aided drug design [1].…”

“…The major problems of the existing methods for predicting protein-protein docking sites are the limitation of existing computation capacity and the high cost of computing resources. 1 Some efforts have used parallelization or graphics processing unit (GPU) to accelerate the calculation, but they require access to a specific type of hardware resource. Due to the flexibility and scalability of computing models, cloud computing has become the main tool to deal with massive data analysis and processing.…”