Assessing the information content of structural and protein–ligand interaction representations for the classification of kinase inhibitor binding modes via machine learning and active learning
Tóm tắt
For kinase inhibitors, X-ray crystallography has revealed different types of binding modes. Currently, more than 2000 kinase inhibitors with known binding modes are available, which makes it possible to derive and test machine learning models for the prediction of inhibitors with different binding modes. We have addressed this prediction task to evaluate and compare the information content of distinct molecular representations including protein–ligand interaction fingerprints (IFPs) and compound structure-based structural fingerprints (i.e., atom environment/fragment fingerprints). IFPs were designed to capture binding mode-specific interaction patterns at different resolution levels. Accurate predictions of kinase inhibitor binding modes were achieved with random forests using both representations. The performance of IFPs was consistently superior to atom environment fingerprints, albeit only by less than 10%. An active learning strategy applying information entropy-based selection of training instances was applied as a diagnostic approach to assess the relative information content of distinct representations. IFPs were found to capture more binding mode-relevant information than atom environment fingerprints, leading to highly predictive models even when training instances were randomly selected. By contrast, for atom environment fingerprints, the derivation of accurate models via active learning depended on entropy-based selection of informative training compounds. Notably, higher information content of IFPs confirmed by active learning only resulted in small improvements in global prediction accuracy compared to models derived using atom environment fingerprints. For practical applications, prediction of binding modes of new kinase inhibitors on the basis of chemical structure is highly attractive.
Tài liệu tham khảo
Klaeger S, Heinzlmeir S, Wilheim M et al (2017) The target landscape of clinical kinase drugs. Science. https://doi.org/10.1126/science.aan4368
Miljković F, Bajorath J (2018) Computational analysis of kinase inhibitors identifies promiscuity cliffs across the human kinome. ACS Omega 3:17295–17308
Hu Y, Bajorath J (2017) Entering the ‘big data’ era in medicinal chemistry: molecular promiscuity analysis revisited. Future Sci 3:FSO179
Miljković F, Bajorath J (2018) Exploring selectivity of multi-kinase inhibitors across the human kinome. ACS Omega 3:1147–1153
Rodríguez-Pérez R, Bajorath J (2019) Multi-task machine learning for classifying highly and weakly potent kinase inhibitors. ACS Omega 4:4367–4375
van Linden OPJ, Kooistra AJ, Leurs R, de Esch IJP, de Graaf C (2014) KLIFS: a knowledge-based structural database to navigate kinase-ligand interaction space. J Med Chem 57:249–277
Kooistra J, Kanev GK, van Linden OPJ et al (2016) KLIFS: a structural kinase-ligand interaction database. Nucleic Acids Res 44:D365–D371
Kalyaanamoorthy S, Chen YP (2011) Structure-based drug design to augment hit discovery. Drug Discov Today 16:831–839
Roskoski R (2016) Classification of small molecule protein kinase inhibitors based upon the structures of their drug-enzyme complexes. Pharmacol Res 103:26–48
Müller S, Chaikuad A, Gray NS, Knapp S (2015) The ins and outs of selective kinase inhibitor development. Nat Chem Biol 11:818–821
Marcou G, Rognan D (2007) Optimizing fragment and scaffold docking by use of molecular interaction fingerprints. J Chem Inf Model 47:195–207
de Graaf C, Kooistra AJ, Vischer HF et al (2011) Crystal structure-based virtual screening for fragment-like ligands of the human histamine H1 receptor. J Med Chem 54:8195–8206
Deng Z, Chuaqui C, Singh J (2004) Structural interaction fingerprint (SIFt): a novel method for analyzing three-dimensional protein-ligand binding interactions. J Med Chem 47:337–344
Rácz A, Bajusz D, Héberger K (2018) Life beyond the Tanimoto coeffcient: similarity measures for interaction fingerprints. J Cheminform 10:48
Da C, Kireev D (2014) Structural Protein-ligand interaction fingerprints (SPLIF) for structure-based virtual screening: method and benchmark study. J Chem Inf Model 54:2555–2561
Kelly MD, Mancera RL (2004) Expanded interaction fingerprint method for analyzing ligand binding modes in docking and structure-based drug design. J Chem Inf Comput Sci 44:1942–1951
Chupakhin V, Marcou G, Baskin I, Varnek A, Rognan D (2013) Predicting ligand binding modes from neural networks trained on protein-ligand interaction fingerprints. J Chem Inf Model 53:763–772
Miljković F, Rodríguez-Pérez R, Bajorath J (2019) Machine learning models for accurate prediction of kinase inhibitors with different binding modes. J Med Chem. https://doi.org/10.1021/acs.jmedchem.9b00867(in press)
Martin E, Mukherjee P (2012) Kinase-kernel models: accurate in silico screening of 4 million compounds across the entire human kinome. J Chem Inf Model 52:763–772
Bosc N, Wroblowski B, Meyer C, Bonnet P (2017) Prediction of protein kinase–ligand interactions through 2.5 D kinochemometrics. J Chem Inf Model 57:93–101
Liu Y, Gray NS (2006) Rational design of inhibitors that bind to inactive kinase conformations. Nat Chem Biol 2:358–364
Zhao Z, Wu H, Wang L et al (2014) Exploration of type II binding mode: a privileged approach for kinase inhibitor focused drug discovery? ACS Chem Biol 9:1230–1241
Rogers D, Hahn M (2010) Extended-connectivity fingerprints. J Chem Inf Model 50:742–754
OEChem TK, version 2.0.0; OpenEye Scientific Software, Santa Fe, NM
Breiman L (2001) Random forests. Mach Learn 45:5–32
Efron B (1979) Bootstrap methods: another look at the Jackknife. Ann Stat 7:1–26
Pedregosa F, Varoquaux G, Gramfort A et al (2011) Scikit-learn: machine learning in Python. J Mach Learn Res 12:2825–2830
Hastie T, Tibshirani R, Friedman JH (2009) The elements of statistical learning. Springer, Berlin
Cohn DA, Ghahramani Z, Jordan MI (1996) Active learning with statistical models. J Artif Intell Res 4:129–145
Shannon CE (1948) A mathematical theory of communication. Bell Labs Tech J 27:379–423
Matthews B (1975) Comparison of the predicted and observed secondary structure of t4 phage lysozyme. Biochim Biophys Acta 405:442–451
Brodersen KH, Ong CS, Stephan KE, Buhmann JM (2010) The balanced accuracy and its posterior distribution. Proceedings of the 20th International Conference on Pattern Recognition (ICPR):3121-3124
Ojala M, Garriga G (2010) Permutation tests for studying classifier performance. J Mach Learn Res 11:1833–1863
Van der Maate L, Hinton G (2008) Visualizing data using t-SNE. J Mach Learn Res 9:2579–2605
Kullback S, Leibler RA (1951) On information and sufficiency. Ann Math Stat 22:79–86
Willet P, Barnard J, Downs G (1998) Chemical similarity searching. J Chem Inf Comp Sci 38:983–996
https://zenodo.org/record/3743636