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.

Computer-aided diagnosis of Alzheimer's disease (AD) is a rapidly developing field of neuroimaging with strong potential to be used in practice. In this context, assessment of models' robustness to noise and imaging protocol differences together with post-processing and tuning strategies are key tasks to be addressed in order to move towards successful clinical applications. In this study, we investigated the efficacy of Random Forest classifiers trained using different structural MRI measures, with and without neuroanatomical constraints in the detection and prediction of AD in terms of accuracy and between-cohort robustness. From The ADNI database, 185 AD, and 225 healthy controls (HC) were randomly split into training and testing datasets. 165 subjects with mild cognitive impairment (MCI) were distributed according to the month of conversion to dementia (4-year follow-up). Structural 1.5-T MRI-scans were processed using Freesurfer segmentation and cortical reconstruction. Using the resulting output, AD/HC classifiers were trained. Training included model tuning and performance assessment using out-of-bag estimation. Subsequently the classifiers were validated on the AD/HC test set and for the ability to predict MCI-to-AD conversion. Models' between-cohort robustness was additionally assessed using the AddNeuroMed dataset acquired with harmonized clinical and imaging protocols. In the ADNI set, the best AD/HC sensitivity/specificity (88.6%/92.0% - test set) was achieved by combining cortical thickness and volumetric measures. The Random Forest model resulted in significantly higher accuracy compared to the reference classifier (linear Support Vector Machine). The models trained using parcelled and high-dimensional (HD) input demonstrated equivalent performance, but the former was more effective in terms of computation/memory and time costs. The sensitivity/specificity for detecting MCI-to-AD conversion (but not AD/HC classification performance) was further improved from 79.5%/75%-83.3%/81.3% by a combination of morphometric measurements with ApoE-genotype and demographics (age, sex, education). When applied to the independent AddNeuroMed cohort, the best ADNI models produced equivalent performance without substantial accuracy drop, suggesting good robustness sufficient for future clinical implementation.

Cyclooxygenases (COX) are present in the body in two isoforms, namely: COX-1, constitutively expressed, and COX-2, induced in physiopathological conditions such as cancer or chronic inflammation. The inhibition of COX with non-steroideal anti-inflammatory drugs (NSAIDs) is the most widely used treatment for chronic inflammation despite the adverse effects associated to prolonged NSAIDs intake. Although selective COX-2 inhibition has been shown not to palliate all adverse effects (e.g. cardiotoxicity), there are still niche populations which can benefit from selective COX-2 inhibition. Thus, capitalizing on bioactivity data from both isoforms simultaneously would contribute to develop COX inhibitors with better safety profiles. We applied ensemble proteochemometric modeling (PCM) for the prediction of the potency of 3,228 distinct COX inhibitors on 11 mammalian cyclooxygenases. Ensemble PCM models ( $R_{0\ test}^{2}=0.65$ R 0 test 2 = 0.65 , and RMSE_test = 0.71) outperformed models exclusively trained on compound ( $R_{0\ test}^{2}=0.17$ R 0 test 2 = 0.17 , and RMSE_test = 1.09) or protein descriptors ( $R_{0\ test}^{2}=0.16$ R 0 test 2 = 0.16 and RMSE_test = 1.10) on the test set. Moreover, PCM predicted COX potency for 1,086 selective and non-selective COX inhibitors with $R_{0\ test}^{2}=0.59$ R 0 test 2 = 0.59 and RMSE_test = 0.76. These values are in agreement with the maximum and minimum achievable $R_{0\ test}^{2}$ R 0 test 2 and RMSE_test values of approximately 0.68 for both metrics. Confidence intervals for individual predictions were calculated from the standard deviation of the predictions from the individual models composing the ensembles. Finally, two substructure analysis pipelines singled out chemical substructures implicated in both potency and selectivity in agreement with the literature.

‘Proteochemometric modeling’ is a bioactivity modeling technique founded on the description of both small molecules (the ligands), and proteins (the targets). By combining those two elements of a ligand – target interaction proteochemometrics techniques model the interaction complex or the full ligand – target interaction space, and they are able to quantify the similarity between both ligands and targets simultaneously. Consequently, proteochemometric models or complex based models, can be considered an extension of QSAR models, which are ligand based. As proteochemometric models are able to incorporate target information they outperform conventional QSAR models when extrapolating from the activities of known ligands on known targets to novel targets. Vice versa, proteochemometrics can be used to virtually screen for selective compounds that are solely active on a single member of a subfamily of targets, as well as to select compounds with a desired bioactivity profile – a topic particularly relevant with concepts such as ‘ligand polypharmacology’ in mind. Here we illustrate the concept of proteochemometrics and provide a review of relevant methodological publications in the field. We give an overview of the target families proteochemometrics modeling has previously been applied to, and introduce some novel application areas of the modeling technique. We conclude that proteochemometrics is a promising technique in preclinical drug research that allows merging data sets that were previously considered separately, with the potential to extrapolate more reliably both in ligand as well as target space.