Dataset Information

Machine-learning scoring functions trained on complexes dissimilar to the test set already outperform classical counterparts on a blind benchmark.

ABSTRACT: The superior performance of machine-learning scoring functions for docking has caused a series of debates on whether it is due to learning knowledge from training data that are similar in some sense to the test data. With a systematically revised methodology and a blind benchmark realistically mimicking the process of prospective prediction of binding affinity, we have evaluated three broadly used classical scoring functions and five machine-learning counterparts calibrated with both random forest and extreme gradient boosting using both solo and hybrid features, showing for the first time that machine-learning scoring functions trained exclusively on a proportion of as low as 8% complexes dissimilar to the test set already outperform classical scoring functions, a percentage that is far lower than what has been recently reported on all the three CASF benchmarks. The performance of machine-learning scoring functions is underestimated due to the absence of similar samples in some artificially created training sets that discard the full spectrum of complexes to be found in a prospective environment. Given the inevitability of any degree of similarity contained in a large dataset, the criteria for scoring function selection depend on which one can make the best use of all available materials. Software code and data are provided at https://github.com/cusdulab/MLSF for interested readers to rapidly rebuild the scoring functions and reproduce our results, even to make extended analyses on their own benchmarks.

SUBMITTER: Li H

PROVIDER: S-EPMC8575004 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:The increase of publicly available bioactivity data in recent years has fueled and catalyzed research in chemogenomics, data mining, and modeling approaches. As a direct result, over the past few years a multitude of different methods have been reported and evaluated, such as target fishing, nearest neighbor similarity-based methods, and Quantitative Structure Activity Relationship (QSAR)-based protocols. However, such studies are typically conducted on different datasets, using different validation strategies, and different metrics. In this study, different methods were compared using one single standardized dataset obtained from ChEMBL, which is made available to the public, using standardized metrics (BEDROC and Matthews Correlation Coefficient). Specifically, the performance of Naïve Bayes, Random Forests, Support Vector Machines, Logistic Regression, and Deep Neural Networks was assessed using QSAR and proteochemometric (PCM) methods. All methods were validated using both a random split validation and a temporal validation, with the latter being a more realistic benchmark of expected prospective execution. Deep Neural Networks are the top performing classifiers, highlighting the added value of Deep Neural Networks over other more conventional methods. Moreover, the best method ('DNN_PCM') performed significantly better at almost one standard deviation higher than the mean performance. Furthermore, Multi-task and PCM implementations were shown to improve performance over single task Deep Neural Networks. Conversely, target prediction performed almost two standard deviations under the mean performance. Random Forests, Support Vector Machines, and Logistic Regression performed around mean performance. Finally, using an ensemble of DNNs, alongside additional tuning, enhanced the relative performance by another 27% (compared with unoptimized 'DNN_PCM'). Here, a standardized set to test and evaluate different machine learning algorithms in the context of multi-task learning is offered by providing the data and the protocols. Graphical Abstract .

Project description:BackgroundMultiple scoring systems have been developed for both the intensive care unit (ICU) and the emergency department (ED) to risk stratify patients and predict mortality. However, it remains unclear whether the additional data needed to compute ICU scores improves mortality prediction for critically ill patients compared to the simpler ED scores.MethodsWe studied a prospective observational cohort of 227 critically ill patients admitted to the ICU directly from the ED at an academic, tertiary care medical center. We compared Acute Physiology and Chronic Health Evaluation (APACHE) II, APACHE III, Simplified Acute Physiology Score (SAPS) II, Modified Early Warning Score (MEWS), Rapid Emergency Medicine Score (REMS), Prince of Wales Emergency Department Score (PEDS), and a pre-hospital critical illness prediction score developed by Seymour et al. (JAMA 2010, 304(7):747-754). The primary endpoint was 60-day mortality. We compared the receiver operating characteristic (ROC) curves of the different scores and their calibration using the Hosmer-Lemeshow goodness-of-fit test and visual assessment.ResultsThe ICU scores outperformed the ED scores with higher area under the curve (AUC) values (p?=?0.01). There were no differences in discrimination among the ED-based scoring systems (AUC 0.698 to 0.742; p?=?0.45) or among the ICU-based scoring systems (AUC 0.779 to 0.799; p?=?0.60). With the exception of the Seymour score, the ED-based scoring systems did not discriminate as well as the best-performing ICU-based scoring system, APACHE III (p?=?0.005 to 0.01 for comparison of ED scores to APACHE III). The Seymour score had a superior AUC to other ED scores and, despite a lower AUC than all the ICU scores, was not significantly different than APACHE III (p?=?0.09). When data from the first 24 h in the ICU was used to calculate the ED scores, the AUC for the ED scores improved numerically, but this improvement was not statistically significant. All scores had acceptable calibration.ConclusionsIn contrast to prior studies of patients based in the emergency department, ICU scores outperformed ED scores in critically ill patients admitted from the emergency department. This difference in performance seemed to be primarily due to the complexity of the scores rather than the time window from which the data was derived.

Project description:We have collected computed barrier heights and reaction energies (and associated model structures) for five enzymes from studies published by Himo and co-workers. Using this data, obtained at the B3LYP/6- 311+G(2d,2p)[LANL2DZ]//B3LYP/6-31G(d,p) level of theory, we then benchmark PM6, PM7, PM7-TS, and DFTB3 and discuss the influence of system size, bulk solvation, and geometry re-optimization on the error. The mean absolute differences (MADs) observed for these five enzyme model systems are similar to those observed for PM6 and PM7 for smaller systems (10-15 kcal/mol), while DFTB results in a MAD that is significantly lower (6 kcal/mol). The MADs for PMx and DFTB3 are each dominated by large errors for a single system and if the system is disregarded the MADs fall to 4-5 kcal/mol. Overall, results for the condensed phase are neither more or less accurate relative to B3LYP than those in the gas phase. With the exception of PM7-TS, the MAD for small and large structural models are very similar, with a maximum deviation of 3 kcal/mol for PM6. Geometry optimization with PM6 shows that for one system this method predicts a different mechanism compared to B3LYP/6-31G(d,p). For the remaining systems, geometry optimization of the large structural model increases the MAD relative to single points, by 2.5 and 1.8 kcal/mol for barriers and reaction energies. For the small structural model, the corresponding MADs decrease by 0.4 and 1.2 kcal/mol, respectively. However, despite these small changes, significant changes in the structures are observed for some systems, such as proton transfer and hydrogen bonding rearrangements. The paper represents the first step in the process of creating a benchmark set of barriers computed for systems that are relatively large and representative of enzymatic reactions, a considerable challenge for any one research group but possible through a concerted effort by the community. We end by outlining steps needed to expand and improve the data set and how other researchers can contribute to the process.

Dataset Information

Machine-learning scoring functions trained on complexes dissimilar to the test set already outperform classical counterparts on a blind benchmark.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets