Data-driven Chemical Reaction Prediction and Retrosynthesis.

The synthesis of organic compounds, which is central to many areas such as drug discovery, material synthesis and biomolecular chemistry, requires chemists to have years of knowledge and experience. The development of technologies with the potential to learn and support experts in the design of synthetic routes is a half-century-old challenge with an interesting revival in the last decade. In fact, the renewed interest in artificial intelligence (AI), driven mainly by data availability, is profoundly changing the landscape of computer-aided chemical reaction prediction and retrosynthetic analysis. In this article, we briefly review different approaches to predict forward reactions and retrosynthesis, with a strong focus on data-driven ones. While data-driven technologies still need to demonstrate their full potential compared to expert rule-based systems in synthetic chemistry, the acceleration experienced in the last decade is a convincing sign that where we use software today, there will be AI tomorrow. This revolution will help and empower bench chemists, driving the transformation of chemistry towards a high-tech business over the next decades.

Inferring experimental procedures from text-based representations of chemical reactions.

The experimental execution of chemical reactions is a context-dependent and time-consuming process, often solved using the experience collected over multiple decades of laboratory work or searching similar, already executed, experimental protocols. Although data-driven schemes, such as retrosynthetic models, are becoming established technologies in synthetic organic chemistry, the conversion of proposed synthetic routes to experimental procedures remains a burden on the shoulder of domain experts. In this work, we present data-driven models for predicting the entire sequence of synthesis steps starting from a textual representation of a chemical equation, for application in batch organic chemistry. We generated a data set of 693,517 chemical equations and associated action sequences by extracting and processing experimental procedure text from patents, using state-of-the-art natural language models. We used the attained data set to train three different models: a nearest-neighbor model based on recently-introduced reaction fingerprints, and two deep-learning sequence-to-sequence models based on the Transformer and BART architectures. An analysis by a trained chemist revealed that the predicted action sequences are adequate for execution without human intervention in more than 50% of the cases.

Prioritizing Delivery of Cancer Treatment During a COVID-19 Lockdown: The Experience of a Clinical Oncology Service in India.

PURPOSE: A COVID-19 lockdown in India posed significant challenges to the continuation of radiotherapy (RT) and systemic therapy services. Although several COVID-19 service guidelines have been promulgated, implementation data are yet unavailable. We performed a comprehensive audit of the implementation of services in a clinical oncology department. METHODS: A departmental protocol of priority-based treatment guidance was developed, and a departmental staff rotation policy was implemented. Data were collected for the period of lockdown on outpatient visits, starting, and delivery of RT and systemic therapy. Adherence to protocol was audited, and factors affecting change from pre-COVID standards analyzed by multivariate logistic regression. RESULTS: Outpatient consults dropped by 58%. Planned RT starts were implemented in 90%, 100%, 92%, 90%, and 75% of priority level 1-5 patients. Although 17% had a deferred start, the median time to start of adjuvant RT and overall treatment times were maintained. Concurrent chemotherapy was administered in 89% of those eligible. Systemic therapy was administered to 84.5% of planned patients. However, 33% and 57% of curative and palliative patients had modifications in cycle duration or deferrals. The patient's inability to come was the most common reason for RT or ST deviation. Factors independently associated with a change from pre-COVID practice was priority-level allocation for RT and age and palliative intent for systemic therapy. CONCLUSION: Despite significant access limitations, a planned priority-based system of delivery of treatment could be implemented.

Automated extraction of chemical synthesis actions from experimental procedures.

Experimental procedures for chemical synthesis are commonly reported in prose in patents or in the scientific literature. The extraction of the details necessary to reproduce and validate a synthesis in a chemical laboratory is often a tedious task requiring extensive human intervention. We present a method to convert unstructured experimental procedures written in English to structured synthetic steps (action sequences) reflecting all the operations needed to successfully conduct the corresponding chemical reactions. To achieve this, we design a set of synthesis actions with predefined properties and a deep-learning sequence to sequence model based on the transformer architecture to convert experimental procedures to action sequences. The model is pretrained on vast amounts of data generated automatically with a custom rule-based natural language processing approach and refined on manually annotated samples. Predictions on our test set result in a perfect (100%) match of the action sequence for 60.8% of sentences, a 90% match for 71.3% of sentences, and a 75% match for 82.4% of sentences.

Predicting retrosynthetic pathways using transformer-based models and a hyper-graph exploration strategy.

We present an extension of our Molecular Transformer model combined with a hyper-graph exploration strategy for automatic retrosynthesis route planning without human intervention. The single-step retrosynthetic model sets a new state of the art for predicting reactants as well as reagents, solvents and catalysts for each retrosynthetic step. We introduce four metrics (coverage, class diversity, round-trip accuracy and Jensen-Shannon divergence) to evaluate the single-step retrosynthetic models, using the forward prediction and a reaction classification model always based on the transformer architecture. The hypergraph is constructed on the fly, and the nodes are filtered and further expanded based on a Bayesian-like probability. We critically assessed the end-to-end framework with several retrosynthesis examples from literature and academic exams. Overall, the frameworks have an excellent performance with few weaknesses related to the training data. The use of the introduced metrics opens up the possibility to optimize entire retrosynthetic frameworks by focusing on the performance of the single-step model only.

Larvicidal activity of Acacia nilotica extracts and isolation of D-pinitol--a bioactive carbohydrate.

A low-molecular-weight, sugar-like compound other than glucose, fructose, sucrose, or myo-inositol showing lipophilic nature was isolated from the EtOH extract of Acacia nilotica. The structure of the compound was determined by spectral methods. This alicyclic polyalcohol was found to be D-pinitol (= 3-O-methyl-D-chiro-inositol; 1). The configuration of the compound was confirmed by single-crystal X-ray analysis. The compound 1 is known from Soybean, Australian mangroves, Fagonia indica, Arachis hypogaea, etc., but we have isolated this compound for the first time from the aerial parts of A. nilotica. Very few references have been cited for compound 1 for its entomological activity, and there are no reports on mosquitoes. Therefore, the crude extracts of A. nilotica were tested for its biological activity against mosquito larvae. Acetone extract at 500-ppm concentration showed chronic toxicity against Aedes aegypti and Culex quinquefasciatus IVth instar mosquito larvae. Such a biological activity has been observed for the first time for this plant. This study could be a stepping stone to a solution for destroying larval species as well as consumption of such a widely grown, problem weed, A. nilotica. This larvicidal agent, since it is derived from plant, is eco-friendly, cost effective, non-hazardous to non-target organisms and would be safe unlike commercially available insecticides.

Did successfully treated pulmonary tuberculosis patients undergo all follow-up sputum smear examinations?

To assess response to anti-tuberculosis treatment as per national guidelines, a retrospective record review was undertaken in four districts of Andhra Pradesh, India, in December 2009 to determine whether pulmonary tuberculosis (PTB) patients reported as successfully treated (cured or treatment completed) underwent all scheduled follow-up sputum smear examinations. In a quarterly cohort of 3000 PTB patients reported as successfully treated, 1847 (61.5%) underwent all follow-up sputum examinations, with a higher proportion of new cases (65%) than retreatment cases (45%). The mid-continuation phase follow-up sputum examinations were commonly missed, and 11% patients had not undergone end-of-treatment follow-up sputum examinations.

On a entrepris en décembre 2009 une revue rétrospective des enregistrements dans quatre districts d'Andhra Pradesh en Inde afin de déterminer dans quelle mesure les patients atteints de tuberculose pulmonaire (TBP) signalés comme traités avec succès (guérison ou achèvement du traitement) avaient subi tous les examens de frottis de crachats prévus dans le suivi pour évaluer la réponse au traitement TB conformément aux directives nationales. Dans une cohorte trimestrielle de 3000 patients TBP signalés comme traités avec succès, 1847 (61,5%) avaient subi des examens de crachats du suivi, la proportion étant plus élevée parmi les nouveaux cas (65%) que parmi les retraitements (45%). Les examens de crachats du suivi au milieu de la phase de continuation ont manqué régulièrement et l'examen de crachats du suivi à la fin du traitement n'a pas été appliqué chez 11% des patients.

En diciembre 2009 se llevó a cabo una revisión retrospectiva de los registros de tuberculosis (TB) en cuatro distritos de Andhra Pradesh en la India, con el fin de determinar si todos los pacientes con TB pulmonar (TBP) cuyo desenlace terapéutico se había notificado como exitoso (curación o tratamiento completo) habían cumplido con las baciloscopias de seguimiento programadas para evaluar la respuesta al tratamiento antituberculoso de conformidad con lo estipulado en las directrices nacionales. En una cohorte trimestral de 3000 pacientes con TBP cuyo tratamiento se registró como exitoso, 1847 pacientes (61,5%) contaban con todas las baciloscopias de seguimiento; esta proporción fue más alta en los casos nuevos (65%) que en los casos de retratamiento (45%). Con frecuencia faltaron las baciloscopias previstas en la mitad de la fase de continuación y 11% de los pacientes carecía del examen de esputo correspondiente a la finalización del tratamiento.

Structure-activity relationship study of androstene steroids with respect to local anti-inflammatory activity.

A sex hormone, 3beta-acetoxy-17beta-hydroxy-androst-5-ene (1) (CAS 1639-43-6), was isolated from aerial parts of Acacia nilotica. This compound is reported to have anti-inflammatory activity. In view of this, considering this molecule as a lead molecule different androstene compounds were synthesized to study their potency and structure-activity relationship with respect to local anti-inflammatory activity. The experiments indicated that 17 ketonic compounds were more active towards inflammation than their oxime analogues. Similarly, for the compounds containing an acetyl group fixed at C-3 position a decreasing trend of activity was observed in the order of ketonic, hydroxyl, oxime and acetyl group, respectively, when these groups are at C-17 position.