Improved Selection of Rare Reactions in Template-Based Retrosynthesis Predictions

Identifying synthetic routes for molecules of interest is a crucial step when discovering new drugs or materials. To find synthetic routes, we can use computer-assisted synthesis planning using expansion policy networks trained on reaction templates extracted from patents and the literature. However, experience has shown that these networks are biased towards frequently reported reactions. This study shows that changing the molecular representation from an extended-connectivity fingerprint to a simple graph representation can increase the accuracy for templates used less than five times by 5.0- 8.5% points. We also illustrate that a simple oversampling of the training set yielded a top-1 accuracy increase in the 17-20% point range for templates used five times or less.

NeuralTPL: a deep learning approach for efficient reaction space exploration

Computer-aided synthesis planning (CASP) has been helping chemists to synthesize novel molecules at an accelerated pace. The recent integration of deep learning with CASP has opened up new avenues for digitizing and exploring the vastly unknown chemical space, and has led to high expectations for fully automated synthesis plannings using machine-discovered novel reactions in the "future". Despite many progresses in the past few years, most deep-learning methods only focus on improving few aspects of CASP (e.g., top-k accuracy). In this work, we target specifically the efficiency of reaction space exploration and its impact on CASP. We propose NeuralTPL, a template-oriented generative approach, that performs impressively across a range of evaluation metrics including chemical validity, diversity, and novelty for various tasks in CASP. In addition, our Transformer-based model bears the potential to learn the core reaction transformation as it can efficiently explore the reaction space. We then perform several experiments and conduct a thorough analysis regarding the three metrics and demonstrate its chemical value for improving the existing deep-learning-driven CASP algorithms.

NeuralTPL: a deep learning approach for efficient reaction space exploration

Computer-aided synthesis planning (CASP) has been helping chemists to synthesize novel molecules at an accelerated pace. The recent integration of deep learning with CASP has opened up new avenues for digitizing and exploring the vastly unknown chemical space, and has led to high expectations for fully automated synthesis plannings using machine-discovered novel reactions in the "future". Despite many progresses in the past few years, most deep-learning methods only focus on improving few aspects of CASP (e.g., top-k accuracy). In this work, we target specifically the efficiency of reaction space exploration and its impact on CASP. We propose NeuralTPL, a template-oriented generative approach, that performs impressively across a range of evaluation metrics including chemical validity, diversity, and novelty for various tasks in CASP. In addition, our Transformer-based model bears the potential to learn the core reaction transformation as it can efficiently explore the reaction space. We then perform several experiments and conduct a thorough analysis regarding the three metrics and demonstrate its chemical value for improving the existing deep-learning-driven CASP algorithms.

A Sequence-to-Sequence Transformer Model for Disconnection Aware Retrosynthesis

Retrosynthesis is an approach commonly undertaken when considering the manufacture of novel molecules. During this process, a target molecule is broken down and analyzed by considering the bonds to be changed as well as the functional group interconversion. In modern computer-assisted synthesis planning tools, the predictions of these changes are typically carried out automatically. However there may be some benefit to the decision being guided by those executing the process: typically, chemists have a clear idea where the retrosynthetic change should happen, but not how such a transformation is to be realized. Using a data-driven model, the retrosynthesis task can be further explored by giving chemists the option to explore specific disconnections. In this work, we design an approach to provide this option by adapting a transformer-based model for single-step retrosynthesis. The model takes as input a product SMILES string, in which the atoms where the transformation should occur are tagged accordingly. This model predicts precursors corresponding to a disconnection occurring in the correct location in 88.9% of the test set reactions. The assessment with a forward prediction model shows that 76% of the predictions are chemically correct, with 14.1% perfectly matching the ground truth.

Paradigm Shift: The Promise of Deep Learning in Molecular Systems Engineering and Design

The application of deep learning to a diverse array of research problems has accelerated progress across many fields, bringing conventional paradigms to a new intelligent era. Just as the roles of instrumentation in the old chemical revolutions, we reinforce the necessity for integrating deep learning in molecular systems engineering and design as a transformative catalyst towards the next chemical revolution. To meet such research needs, we summarize advances and progress across several key elements of molecular systems: molecular representation, property estimation, representation learning, and synthesis planning. We further spotlight recent advances and promising directions for several deep learning architectures, methods, and optimization platforms. Our perspective is of interest to both computational and experimental researchers as it aims to chart a path forward for cross-disciplinary collaborations on synthesizing knowledge from available chemical data and guiding experimental efforts.

Desain Tapak Perencanaan PenangkaranKupu-Kupu di Kampus IPB Darmaga Menggunakan Sistem Zonasi untuk Mendukung Aktivitas Wisata Alam

Butterfly captivity can be successful if there are artificial environmental conditions that are suitable for the life and reproduction of butterflies. Apart from the technical side, the success key of the captivity also determined from the side of site plan. Site planning is the ability to collect and interpreting data, project into the future, identify problems and provide a reasoned approach to solving existing problems. The purpose of this study was to design a captive development site for the IPB Dramaga butterfly captivity based on the conditions and characteristics of the area. The method used in this research consists of literature study, interview and observation. Data analysis consisted of six stages, consists of: preparation, inventory, analysis, synthesis, planning and design. The results of this study were the IPB Dramaga butterfly captivity site was designed into three zoning systems according to the needs of captive management, namely an office zone (0.37 ha), breeding zone (1.75 ha) and a tourism zone (2.13 ha).

Application of Artificial Intelligence to Address Issues Related to the COVID-19 Virus

Artificial intelligence (AI) plays a major role in addressing novel coronavirus 2019 (COVID-19)-related issues and is also used in computer-aided synthesis planning (CASP). AI, including machine learning, is used by artificial neural networks such as deep neural networks and recurrent networks. AI has been used in activity predictions like physicochemical properties. Machine learning in de novo design explores the generation of fruitful, biologically active molecules toward expected or finished products. Several examples establish the strength of machine learning or AI in this field. AI techniques can significantly improve treatment consistency and decision making by developing useful algorithms. AI is helpful not only in the treatment of COVID-19-infected patients but also for their proper health monitoring. It can track the crisis of COVID-19 at different scales, such as medical, molecular, and epidemiological applications. It is also helpful to facilitate the research on this virus by analyzing the available data. AI can help in developing proper treatment regimens, prevention strategies, and drug and vaccine development. Combination with synthesis planning and ease of synthesis are feasible, and more and more automated drug discovery by computers is expected in the near future to eradicate the COVID-19 virus.