Deep Transferable Compound Representation across Domains and Tasks for Low Data Drug Discovery

Abstract

The main problem of small molecule-based drug discovery is to find a candidate molecule with increased pharmacological activity, proper ADME, and low toxicity. Recently, machine learning has driven a significant contribution to drug discovery. However, many machine learning methods, such as deep learning-based approaches, require a large amount of training data to form accurate predictions for unseen data. In lead optimization step, the amount of available biological data on small molecule compounds is low, which makes it a challenging problem to apply machine learning methods. The main goal of this study is to design a new approach to handle these situations. To this end, source assay (auxiliary assay) knowledge is utilized to learn a better model to predict the property of new compounds in the target assay. Up to now, the current approaches did not consider that source and target assays are adapted to different target groups with different compounds distribution. In this paper, we propose a new architecture by utilizing graph convolutional network and adversarial domain adaptation network to tackle this issue. To evaluate the proposed approach, we applied it to Tox21, ToxCast, SIDER, HIV, and BACE collections. The results showed the effectiveness of the proposed approach in transferring the related knowledge from source to target data set.

Artificial Intelligence in Drug Discovery: A Comprehensive Review of Data-driven and Machine Learning Approaches

“…Unlike other ADMET properties, toxicity data is relatively well-known to the public because of the Tox21 challenge in 2014. Therefore, methodology-based studies have recently emerged, and benchmarked against major toxicity datasets including the Tox21 dataset [76,[184][185][186][187]. All of the studies proposed deep learning-based novel architectures to predict the compound toxic-related properties.…”

Synthetic Repurposing of Drugs in Hypertension: a Datamining Method Based on Association Rules and a Novel Discrete Algorithm

“…Overall, three types of machine learning methods, including supervised, semi-supervised, and unsupervised techniques have been used in the scope of drug discovery [15]. Also, it has been shown that some modified and improved versions of the present approaches such as deep neural networks can yield better predictive models [16]. Overfitting and insufficient amounts of data are the main challenges in generating an appropriate predictive model [17,18].…”

Synthetic Repurposing of Drugs in Hypertension: a Datamining Method Based on Association Rules and a Novel Discrete Algorithm

“…Overall, three types of machine learning methods such as supervised, semi-supervised, and unsupervised techniques have been used in drug discovery scope (15). Also, it has been shown that some modified and improved versions of the approaches such as deep neural networks can lead to better predictive models (16). Overfitting and lack of enough number of data are the main challenges in generating an appropriate predictive model (17,18).…”