Performance Comparison of Machine Learning Methods in Discovery of BACE-1 Inhibitors in Alzheimer's Disease Therapy

Abstract

Alzheimer's disease (AD) poses a significant challenge in the realm of neurodegenerative disorders, necessitating effective therapeutic interventions. One promising approach involves the discovery of β-secretase 1 (BACE-1) inhibitors, pivotal in mitigating amyloid-β peptide accumulation, a hallmark of AD pathology. In this study, we compare the performance of three prominent machine learning methods, namely Gradient Boosting Machine (GBM), Random Forest (RF), and Support Vector Machine (SVM) in the discovery of BACE-1 inhibitors. Leveraging the BACE dataset sourced from MoleculeNet, comprising quantitative and qualitative binding results of compounds, we explored the classification efficacy of these methods. Our experimental results reveal distinct precision, recall, and accuracy metrics for each method, showcasing RF with precision and accuracy scores of 1.00 and 99.67%, respectively, followed by GBM and SVM. Furthermore, feature importance analysis underscores pIC50 as the most influential attribute across all methods, emphasizing its pivotal role in classifying BACE-1 inhibitors. Additionally, RF prioritizes Estate as the second most important feature, while AlogP emerges as GBM's secondary significant attribute. These findings shed light on the efficacy of machine learning techniques in identifying potential therapeutics for AD, offering insights into feature importance variations among methods and highlighting the significance of diverse molecular descriptors in drug discovery.

Keywords

• Drug Interaction
• Alzheimer Disease
• BACE-1 inhibitors
• Machine Learning

References

1. [1] M. Thambisetty, L. Beason-Held, M. Kraut, R. Desikan, S. Resnick, & Y. “An. The Entorhinal Cortex-Hippocampal System Is An Early Target Of Clusterin-Related Neurodegeneration In Alzheimer’s Disease”. Alzheimer’s & Dementia, 10(4), Supplement, P160. doi: 10.1016/j.jalz.2014.04.145.
2. [2] E. Ruether, H. Moessler, & M. Windisch. “The MAD-B study — A randomized, double-blind, placebo-controlled trial with cerebrolysin in Alzheimer’s disease”. European Neuropsychopharmacology, 10, 355, 2000, doi: 10.1016/S0924-977X(00)80463-0.
3. [3] D. Al-Jumeily, S. Iram, F. Vialatte, & P. Fergus. “A novel method to analyze EEG synchrony for the early diagnosis of Alzheimer's disease in optimized frequency bands”. In 2014 IEEE 11th Consumer Communications and Networking Conference (CCNC) (pp. 1-4). Las Vegas, NV, USA. doi: 10.1109/CCNC.2014.6866646.
4. [4] M. M. Mishra & P. Kumar. “Crocin: A Potent Secondary Metabolite As BACE1 Inhibitor In Alzheimer’s Disease”. 2023 9th International Conference on Advanced Computing and Communication Systems (ICACCS), s. 1486-1490. Coimbatore, Hindistan. doi: 10.1109/ICACCS57279.2023.10112776.
5. [5] S. Sharma ve Y. Hasija. “Identification and screening of BACE1 inhibitors using Drug Repurposing: A Computational Approach”. 2023 3rd International Conference on Innovative Sustainable Computational Technologies (CISCT), pp. 1-5. Dehradun, Hindistan. doi: 10.1109/CISCT57197.2023.10351386.
6. [6] A. F. Nugroho, R. R. Septiawan, ve I. Kurniawan. “Prediction of Human β-secretase 1 (BACE-1) Inhibitors for Alzheimer Therapeutic Agent by Using Fingerprint-based Neural Network Optimized by Bat Algorithm”. 2023 International Conference on Computer Science, Information Technology and Engineering (ICCoSITE), pp. 257-261. Jakarta, Endonezya. doi: 10.1109/ICCoSITE57641.2023.10127718.
7. [7] Das, B., Yan, R. A Close Look at BACE1 Inhibitors for Alzheimer’s Disease Treatment. CNS Drugs 33, 251–263 (2019). https://doi.org/10.1007/s40263-019-00613-7
8. [8] F. H. Bazzari and A. H. Bazzari, “BACE1 Inhibitors for Alzheimer’s Disease: The Past, Present and Any Future?,” Molecules, vol. 27, no. 24, Art. no. 24, Jan. 2022, doi: 10.3390/molecules27248823.

9. [9] Z. Wang et al. “Inhibitory effects of β-asarone on lncRNA BACE1-mediated induction of autophagy in a model of Alzheimer’s disease. Behavioural Brain Research, vol 463, pp. 114896, 2024.
10. [10] Z. Chang et al. “Signal-on' electrochemical detection of BACE1 for early detection of Alzheimer’s disease”. Cell Reports Physical Science, pp. 101632, 2023.
11. [11] K. Kalaimathi et al. “Cyanobacterial metabolites as novel inhibitors of BACE1 implicated in Alzheimer’s disease through in silico approaches”. Intelligent Pharmacy, vol. 2, no. 1, pp. 144–149, 2024.
12. [12] B. Louis, V. K. Agrawal, ve P. V. Khadikar. “Single crystal X-ray, DFT, molecular dynamic simulations, and biological evaluation of 3-OH pyrrolidine derivative VA10 from alkaloid vasicine for BACE1 inhibition”. Journal of Molecular Structure, vol. 1300, pp. 137196, 2024.
13. [13] M. Nakano, T. Tsuchida, Y. Mitsuishi, ve M. Nishimura. “Nicotinic acetylcholine receptor activation induces BACE1 transcription via the phosphorylation and stabilization of nuclear SP1”. Neuroscience Research, 2023.
14. [14] C. B. Abraham, L. Xu, G. A. Pantelopulos, ve J. E. Straub. “Characterizing the transmembrane domains of ADAM10 and BACE1 and the impact of membrane composition”. Biophysical Journal, cilt. 122, no. 19, pp. 3999–4010,2023.
15. [15] B. Vincent ve S. Maitra. “BACE1-dependent metabolism of neuregulin 1: Bridging the gap in explaining the occurrence of schizophrenia-like symptoms in Alzheimer’s disease with psychosis?”, Ageing Research Reviews, vol. 89, pp. 101988. doi: 10.1016/j.arr.2023.101988, 2023.
16. [16] K. Dominko et. al. “Amyloid-ß plaque formation and BACE1 accumulation in the brains of a 5xFAD Alzheimer’s disease mouse model is associated with altered distribution and not proteolysis of BACE1 substrates Sez6 and Sez6L”. Mechanisms of Ageing and Development, vol. 207, pp. 111726. doi: 10.1016/j.mad.2022.111726, 2022.
17. [17] Z. Wu et. al. “MoleculeNet: A Benchmark for Molecular Machine Learning”. arXiv preprint, arXiv: 1703.00564, 2017.
18. [18] M. M. Inuwa and R. Das, “A comparative analysis of various machine learning methods for anomaly detection in cyber attacks on IoT networks,” Internet of Things, vol. 26, p. 101162, Jul. 2024, doi: 10.1016/j.iot.2024.101162.
19. [19] J. Lin et. al. “Accurate prediction of potential druggable proteins based on genetic algorithm and Bagging-SVM ensemble classifier”. Artificial Intelligence in Medicine, vol. 98, pp. 35–47. doi: 10.1016/j.artmed.2019.07.005, 2019
20. [20] J. I. B. Janairo. “Chapter 6 - Support vector machine in drug design. In Cheminformatics, QSAR and Machine Learning Applications for Novel Drug Development”, K. Roy (Ed.), Academic Press, pp. 161–179. doi: 10.1016/B978-0-443-18638-7.00021-9, 2023.
21. [21] B. Das, S. Toraman, I. Turkoglu, "A novel genome analysis method with the entropy-based numerical techniqueusing pretrained convolutional neural networks," Turkish Journal of Electrical Engineering and Computer Sciences: Vol. 28: No. 4, Article 9, 2020. https://doi.org/10.3906/elk-1909-119
22. [22] A. F. Amiri, H. Oudira, A. Chouder, ve S. Kichou. “Faults detection and diagnosis of PV systems based on machine learning approach using random forest classifier”. Energy Conversion and Management, vol. 301, pp. 118076, 2024, doi: 10.1016/j.enconman.2024.118076.
23. [23] A. Alloubani, B. Abuhaija, M. Almatari, G. Jaradat, ve B. Ihnaini. “Predicting vitamin D deficiency using optimized random forest classifier”. Clinical Nutrition ESPEN, vol. 60, pp. 1–10, 2024. doi: 10.1016/j.clnesp.2023.12.146.
24. [24] G. Cano et. al. “Automatic selection of molecular descriptors using random forest: Application to drug discovery”. Expert Systems with Applications, vol. 72, pp. 151–159, 2017. doi: 10.1016/j.eswa.2016.12.008.
25. [25] Das B. An implementation of a hybrid method based on machine learning to identify biomarkers in the Covid-19 diagnosis using DNA sequences. Chemometr Intell Lab Syst. 2022 Nov 15;230:104680. doi: 10.1016/j.chemolab.2022.104680.
26. [26] P. B. Yang, Y. J. Chan, S. K. Yazdi, and J. W. Lim, “Optimisation and economic analysis of industrial-scale anaerobic co-digestion (ACoD) of palm oil mill effluent (POME) and decanter cake (DC) using machine learning models: A comparative study of Gradient Boosting Machines (GBM), K-nearest neighbours (KNN), and random forest (RF),” Journal of Water Process Engineering, vol. 58, p. 104752, Feb. 2024, doi: 10.1016/j.jwpe.2023.104752.
27. [27] P. B. Yang, Y. J. Chan, S. K. Yazdi, and J. W. Lim, “Optimisation and economic analysis of industrial-scale anaerobic co-digestion (ACoD) of palm oil mill effluent (POME) and decanter cake (DC) using machine learning models: A comparative study of Gradient Boosting Machines (GBM), K-nearest neighbours (KNN), and random forest (RF),” Journal of Water Process Engineering, vol. 58, p. 104752, Feb. 2024, doi: 10.1016/j.jwpe.2023.104752.