DrugEx v2: de novo design of drug molecules by Pareto-based multi-objective reinforcement learning in polypharmacology

In polypharmacology drugs are required to bind to multiple specific targets, for example to enhance efficacy or to reduce resistance formation. Although deep learning has achieved a breakthrough in de novo design in drug discovery, most of its applications only focus on a single drug target to generate drug-like active molecules. However, in reality drug molecules often interact with more than one target which can have desired (polypharmacology) or undesired (toxicity) effects. In a previous study we proposed a new method named DrugEx that integrates an exploration strategy into RNN-based reinforcement learning to improve the diversity of the generated molecules. Here, we extended our DrugEx algorithm with multi-objective optimization to generate drug-like molecules towards multiple targets or one specific target while avoiding off-targets (the two adenosine receptors, A1AR and A2AAR, and the potassium ion channel hERG in this study). In our model, we applied an RNN as the agent and machine learning predictors as the environment. Both the agent and the environment were pre-trained in advance and then interplayed under a reinforcement learning framework. The concept of evolutionary algorithms was merged into our method such that crossover and mutation operations were implemented by the same deep learning model as the agent. During the training loop, the agent generates a batch of SMILES-based molecules. Subsequently scores for all objectives provided by the environment are used to construct Pareto ranks of the generated molecules. For this ranking a non-dominated sorting algorithm and a Tanimoto-based crowding distance algorithm using chemical fingerprints are applied. Here, we adopted GPU acceleration to speed up the process of Pareto optimization. The final reward of each molecule is calculated based on the Pareto ranking with the ranking selection algorithm. The agent is trained under the guidance of the reward to make sure it can generate desired molecules after convergence of the training process. All in all we demonstrate generation of compounds with a diverse predicted selectivity profile towards multiple targets, offering the potential of high efficacy and low toxicity.

DrugEx v2: De Novo Design of Drug Molecule by Pareto-based Multi-Objective Reinforcement Learning in Polypharmacology

<p>In polypharmacology, ideal drugs are required to bind to multiple specific targets to enhance efficacy or to reduce resistance formation. Although deep learning has achieved breakthrough in drug discovery, most of its applications only focus on a single drug target to generate drug-like active molecules in spite of the reality that drug molecules often interact with more than one target which can have desired (polypharmacology) or undesired (toxicity) effects. In a previous study we proposed a new method named <i>DrugEx</i> that integrates an exploration strategy into RNN-based reinforcement learning to improve the diversity of the generated molecules. Here, we extended our <i>DrugEx</i> algorithm with multi-objective optimization to generate drug molecules towards more than one specific target (two adenosine receptors, A₁AR and A_2AAR, and the potassium ion channel hERG in this study). In our model, we applied an RNN as the <i>agent</i> and machine learning predictors as the <i>environment</i>, both of which were pre-trained in advance and then interplayed under the reinforcement learning framework. The concept of evolutionary algorithms was merged into our method such that <i>crossover</i> and <i>mutation</i> operations were implemented by the same deep learning model as the <i>agent</i>. During the training loop, the agent generates a batch of SMILES-based molecules. Subsequently scores for all objectives provided by the <i>environment</i> are used for constructing Pareto ranks of the generated molecules with non-dominated sorting and Tanimoto-based crowding distance algorithms. Here, we adopted GPU acceleration to speed up the process of Pareto optimization. The final reward of each molecule is calculated based on the Pareto ranking with the ranking selection algorithm. The agent is trained under the guidance of the reward to make sure it can generate more desired molecules after convergence of the training process. All in all we demonstrate generation of compounds with a diverse predicted selectivity profile toward multiple targets, offering the potential of high efficacy and lower toxicity.</p>

DrugEx v2: De Novo Design of Drug Molecule by Pareto-based Multi-Objective Reinforcement Learning in Polypharmacology

<p>In polypharmacology, ideal drugs are required to bind to multiple specific targets to enhance efficacy or to reduce resistance formation. Although deep learning has achieved breakthrough in drug discovery, most of its applications only focus on a single drug target to generate drug-like active molecules in spite of the reality that drug molecules often interact with more than one target which can have desired (polypharmacology) or undesired (toxicity) effects. In a previous study we proposed a new method named <i>DrugEx</i> that integrates an exploration strategy into RNN-based reinforcement learning to improve the diversity of the generated molecules. Here, we extended our <i>DrugEx</i> algorithm with multi-objective optimization to generate drug molecules towards more than one specific target (two adenosine receptors, A₁AR and A_2AAR, and the potassium ion channel hERG in this study). In our model, we applied an RNN as the <i>agent</i> and machine learning predictors as the <i>environment</i>, both of which were pre-trained in advance and then interplayed under the reinforcement learning framework. The concept of evolutionary algorithms was merged into our method such that <i>crossover</i> and <i>mutation</i> operations were implemented by the same deep learning model as the <i>agent</i>. During the training loop, the agent generates a batch of SMILES-based molecules. Subsequently scores for all objectives provided by the <i>environment</i> are used for constructing Pareto ranks of the generated molecules with non-dominated sorting and Tanimoto-based crowding distance algorithms. Here, we adopted GPU acceleration to speed up the process of Pareto optimization. The final reward of each molecule is calculated based on the Pareto ranking with the ranking selection algorithm. The agent is trained under the guidance of the reward to make sure it can generate more desired molecules after convergence of the training process. All in all we demonstrate generation of compounds with a diverse predicted selectivity profile toward multiple targets, offering the potential of high efficacy and lower toxicity.</p>

DrugEx v2: De Novo Design of Drug Molecule by Pareto-based Multi-Objective Reinforcement Learning in Polypharmacology

<p>In polypharmacology, ideal drugs are required to bind to multiple specific targets to enhance efficacy or to reduce resistance formation. Although deep learning has achieved breakthrough in drug discovery, most of its applications only focus on a single drug target to generate drug-like active molecules in spite of the reality that drug molecules often interact with more than one target which can have desired (polypharmacology) or undesired (toxicity) effects. In a previous study we proposed a new method named <i>DrugEx</i> that integrates an exploration strategy into RNN-based reinforcement learning to improve the diversity of the generated molecules. Here, we extended our <i>DrugEx</i> algorithm with multi-objective optimization to generate drug molecules towards more than one specific target (two adenosine receptors, A₁AR and A_2AAR, and the potassium ion channel hERG in this study). In our model, we applied an RNN as the <i>agent</i> and machine learning predictors as the <i>environment</i>, both of which were pre-trained in advance and then interplayed under the reinforcement learning framework. The concept of evolutionary algorithms was merged into our method such that <i>crossover</i> and <i>mutation</i> operations were implemented by the same deep learning model as the <i>agent</i>. During the training loop, the agent generates a batch of SMILES-based molecules. Subsequently scores for all objectives provided by the <i>environment</i> are used for constructing Pareto ranks of the generated molecules with non-dominated sorting and Tanimoto-based crowding distance algorithms. Here, we adopted GPU acceleration to speed up the process of Pareto optimization. The final reward of each molecule is calculated based on the Pareto ranking with the ranking selection algorithm. The agent is trained under the guidance of the reward to make sure it can generate more desired molecules after convergence of the training process. All in all we demonstrate generation of compounds with a diverse predicted selectivity profile toward multiple targets, offering the potential of high efficacy and lower toxicity.</p>

Design of Experiment Centered Multivariate Optimization Strategy for RP-HPLC Method to Quantitate Retapamulin in Topical Cream and Microemulsion

Background: Retapamulin is the first pleuromutilin antibacterial approved for the treatment of impetigo. The objective of the current research was to utilize the design of experiments approach for development and optimization of robust RP-HPLC method for the quantitation of Retapamulin in marketed cream and in-house developed microemulsion based formulations with an oily matrix. Methods: The impact of various chromatographic conditions (independent variables) was assessed using Plackett–Burman design on critical analytical attributes (response) to screen initial experimental conditions. The Box-Behnken design was employed to optimize the selected chromatographic factors on the responses. Further, validation of optimized RP-HPLC was carried out as per the ICHQ2(R1) guideline. Results: Pareto ranking analysis showed that % organic phase, flow rate, and volume of injection were found statistically significant (p < 0.05) variables influencing the retention time, number of plates, and tailing of the Retapamulin peak. The optimized RP-HPLC method with the stationary phase, C18 (250 mm × 4.6 mm, 5 μm) column, and mobile phase as a mixture of methanol and potassium dihydrogen orthophosphate buffer (50 mM, pH 7.0, 90:10 % v/v, isocratic), the flow rate of 1.0 mL/min, 10 μL injection volume, 25°C column oven temperature, 247 nm as detection wavelength, was successfully validated based on ICHQ2(R1) guideline. Conclusion: RP-HPLC method was successfully used to separate (retention time 4.34 ± 0.2 min)and assay Retapamulin in microemulsion and marketed cream. The outcomes of the investigation exhibited the effective application of a multivariant approach in the optimization of the RP-HPLCfor routine analysis of Retapamulin.

Deep Open-Set Domain Adaptation for Cross-Scene Classification based on Adversarial Learning and Pareto Ranking

Most of the existing domain adaptation (DA) methods proposed in the context of remote sensing imagery assume the presence of the same land-cover classes in the source and target domains. Yet, this assumption is not always realistic in practice as the target domain may contain additional classes unknown to the source leading to the so-called open set DA. Under this challenging setting, the problem turns to reducing the distribution discrepancy between the shared classes in both domains besides the detection of the unknown class samples in the target domain. To deal with the openset problem, we propose an approach based on adversarial learning and pareto-based ranking. In particular, the method leverages the distribution discrepancy between the source and target domains using min-max entropy optimization. During the alignment process, it identifies candidate samples of the unknown class from the target domain through a pareto-based ranking scheme that uses ambiguity criteria based on entropy and the distance to source class prototype. Promising results using two cross-domain datasets that consist of very high resolution and extremely high resolution images, show the effectiveness of the proposed method.

Gender-Based Vulnerability: Combining Pareto ranking and geostatistics to model gender-based vulnerability in Rohingya refugee settlements in Bangladesh

Background: The Rohingya refugee crisis in Bangladesh continues to outstrip humanitarian resources and undermine the health and security of over 900,000 people. Spatial, sector-specific information is required to better understand the needs of vulnerable populations, such as women and girls, and to target interventions with improved efficiency and effectiveness. This study aimed to create a gender-based vulnerability index and explore the geospatial and thematic variations in gender-based vulnerability of Rohingya refugees residing in Bangladesh by utilizing pre-existing, open source data. Methods: Data sources included remotely-sensed REACH data on humanitarian infrastructure, United Nations Population Fund resource availability data, and the Needs and Population Monitoring Survey conducted by the International Organization for Migration in October 2017. Data gaps were addressed through probabilistic interpolation. A vulnerability index was designed through a process of literature review, variable selection and thematic grouping, normalization, and scorecard creation, and Pareto ranking was employed to rank sites based on vulnerability scoring. Spatial autocorrelation of vulnerability was analyzed with the Global and Anselin Local Moran’s I applied to both combined vulnerability index rank and disaggregated thematic ranking. Results: Twenty-four point one percent of settlements were ranked as ‘most vulnerable,’ with 30 highly vulnerable clusters identified predominantly in the upazila of Sadar. Five settlements in Dhokkin, Somitapara, and Pahartoli were categorized as less vulnerable outliers amongst highly vulnerable neighboring sites. Security- and health-related variables appear to be the most significant drivers of gender-specific vulnerability in Cox’s Bazar. Clusters of low security and education vulnerability measures are shown near Kutupalong. Conclusion: The humanitarian sector produces tremendous amounts of data that can be analyzed with spatial statistics to improve research targeting and programmatic intervention. The critical utilization of these data and the validation of vulnerability indexes are required to improve the international response to the global refugee crisis. This study presents a novel methodology that can be utilized to not only spatially characterize gender-based vulnerability in refugee populations, but can also be calibrated to identify and serve other vulnerable populations during crises.