Chemical Representation Research Articles

Comparing SMILES and SELFIES tokenization for enhanced chemical language modeling

Life sciences research and experimentation are resource-intensive, requiring extensive trials and considerable time. Often, experiments do not achieve their intended objectives, but progress is made through trial and error, eventually leading to breakthroughs. Machine learning is transforming this traditional approach, providing methods to expedite processes and accelerate discoveries. Deep Learning is becoming increasingly prominent in chemistry, with Convolutional Graph Networks (CGN) being a key focus, though other approaches also show significant potential. This research explores the application of Natural Language Processing (NLP) to evaluate the effectiveness of chemical language representations, specifically SMILES and SELFIES, using tokenization methods such as Byte Pair Encoding (BPE) and a novel approach developed in this study, Atom Pair Encoding (APE), in BERT-based models. The primary objective is to assess how these tokenization techniques influence the performance of chemical language models in biophysics and physiology classification tasks. The findings reveal that APE, particularly when used with SMILES representations, significantly outperforms BPE by preserving the integrity and contextual relationships among chemical elements, thereby enhancing classification accuracy. Performance was evaluated in downstream classification tasks using three distinct datasets for HIV, toxicology, and blood–brain barrier penetration, with ROC-AUC serving as the evaluation metric. This study highlights the critical role of tokenization in processing chemical language and suggests that refining these techniques could lead to significant advancements in drug discovery and material science.

Cheminformatic Identification of Tyrosyl-DNA Phosphodiesterase 1 (Tdp1) Inhibitors: A Comparative Study of SMILES-Based Supervised Machine Learning Models.

Tyrosyl-DNA phosphodiesterase 1 (Tdp1) repairs damages in DNA induced by abortive topoisomerase 1 activity; however, maintenance of genetic integrity may sustain cellular division of neoplastic cells. It follows that Tdp1-targeting chemical inhibitors could synergize well with existing chemotherapy drugs to deny cancer growth; therefore, identification of Tdp1 inhibitors may advance precision medicine in oncology. Current computational research efforts focus primarily on molecular docking simulations, though datasets involving three-dimensional molecular structures are often hard to curate and computationally expensive to store and process. We propose the use of simplified molecular input line entry system (SMILES) chemical representations to train supervised machine learning (ML) models, aiming to predict potential Tdp1 inhibitors. An open-sourced consensus dataset containing the inhibitory activity of numerous chemicals against Tdp1 was obtained from Kaggle. Various ML algorithms were trained, ranging from simple algorithms to ensemble methods and deep neural networks. For algorithms requiring numerical data, SMILES were converted to chemical descriptors using RDKit, an open-sourced Python cheminformatics library. Out of 13 optimized ML models with rigorously tuned hyperparameters, the random forest model gave the best results, yielding a receiver operating characteristics-area under curve of 0.7421, testing accuracy of 0.6815, sensitivity of 0.6444, specificity of 0.7156, precision of 0.6753, and F1 score of 0.6595. Ensemble methods, especially the bootstrap aggregation mechanism adopted by random forest, outperformed other ML algorithms in classifying Tdp1 inhibitors from non-inhibitors using SMILES. The discovery of Tdp1 inhibitors could unlock more treatment regimens for cancer patients, allowing for therapies tailored to the patient's condition.

Chemoinformatic Characterization of NAPROC-13: A Database for Natural Product 13C NMR Dereplication.

Natural products (NPs) are secondary metabolites of natural origin with broad applications across various human activities, particularly the discovery of bioactive compounds. Structural elucidation of new NPs entails significant cost and effort. On the other hand, the dereplication of known compounds is crucial for the early exclusion of irrelevant compounds in contemporary pharmaceutical research. NAPROC-13 stands out as a publicly accessible database, providing structural and 13C NMR spectroscopic information for over 25 000 compounds, rendering it a pivotal resource in natural product (NP) research, favoring open science. This study seeks to quantitatively analyze the chemical content, structural diversity, and chemical space coverage of NPs within NAPROC-13, compared to FDA-approved drugs and a very diverse subset of NPs, UNPD-A. Findings indicated that NPs in NAPROC-13 exhibit properties comparable to those in UNPD-A, albeit showcasing a notably diverse array of structural content, scaffolds, ring systems of pharmaceutical interest, and molecular fragments. NAPROC-13 covers a specific region of the chemical multiverse (a generalization of the chemical space from different chemical representations) regarding physicochemical properties and a region as broad as UNPD-A in terms of the structural features represented by fingerprints.

Effect of Computer Simulations on Student Ability to Translate Chemical Representations When Learning the Particulate Nature of Matter Concept

Effect of Computer Simulations on Student Ability to Translate Chemical Representations When Learning the Particulate Nature of Matter Concept

Green chemistry for all: three principles of Inclusive Green and Sustainable Chemistry Education

Abstract The three principles of Inclusive Green and Sustainable Chemistry Education (IGSCE) are presented to guide the reflection, design, and implementation of potentially inclusive materials and approaches. These principles refer to (i) embracing student-centered learning, (ii) promoting teaching in the five levels of representation in chemistry, and (iii) adapting the curriculum to empower students to apply their academic skills effectively to real-life situations through supportive teaching and social guidance. Educational elements conducive to potentially inclusive classrooms and their interconnections are identified and discussed. These include using the Triangular Bipyramid Metaphor (TBM) to facilitate academic inclusivity for individuals with and without disabilities, such as those who are deaf and blind. Further, the importance of ensuring that all students, regardless of their abilities, can fully participate in the educational experience is highlighted, aligning with Sustainable Development Goal 4 (SDG #4) to achieve inclusive education and lifelong learning opportunities. Green chemistry should be available to everyone, not just a few. It promotes sustainable development and deserves global recognition and support. The change agents targeted by these three principles of IGSCE include, but are not limited to, educators, researchers, teachers, and students in secondary and university education.

MPOCSR: optical chemical structure recognition based on multi-path Vision Transformer

AbstractOptical chemical structure recognition (OCSR) is a fundamental and crucial task in the field of chemistry, which aims at transforming intricate chemical structure images into machine-readable formats. Current deep learning-based OCSR methods typically use image feature extractors to extract visual features and employ encoder-decoder architectures for chemical structure recognition. However, the performance of these methods is limited by their image feature extractors and the class imbalance of elements in chemical structure representation. This paper proposes MPOCSR (multi-path optical chemical structure recognition), which introduces the multi-path Vision Transformer (MPViT) and the class-balanced (CB) loss function to address these two challenges. MPOCSR uses MPViT as an image feature extractor, combining the advantages of convolutional neural networks and Vision Transformers. This strategy enables the provision of richer visual information for subsequent decoding processes. Furthermore, MPOCSR incorporates CB loss function to rebalance the loss weights among different categories. For training and validation of our method, we constructed a dataset that includes both Markush and non-Markush structures. Experimental results show that MPOCSR achieves an accuracy of 90.95% on the test set, surpassing other existing methods.

Enhancing Conceptual Understanding of Buffer Solutions with an Intertextual E-Book Prototype

<p>The intertextual-based e-book prototype helps students understand buffer solutions by connecting the three levels of chemical representation: macroscopic, submicroscopic, and symbolic. This connection allows students to visualize chemical phenomena from simple to complex. The study aims to enhance students' conceptual understanding using this e-book prototype. The research and development (R&D) method followed five stages: research and information gathering, product development planning, initial product development, limited trials, and initial product revision. This process yielded a highly validated e-book prototype that promotes innovation. The validation results showed high feasibility in substance (85.71%), instructional methods (90.90%), and language (100%). Experts deemed the e-book prototype highly effective in improving conceptual understanding through its intertextual approach. Pretest and posttest assessments of 30 students indicated improved conceptual understanding after using the e-book. Initially, students demonstrated macroscopic and symbolic level abilities without associating the three levels of representation. After using the e-book, they could connect these levels, enhancing their understanding and conceptual understanding. The N_Gain scores for pretest and posttest assessments were 0.48 for the high category, 0.45 for the medium category, and 0.44 for the low category, showing moderate improvement. The t-test results (Sig. (2-tailed) = 0.000 < 0.05) confirmed the effectiveness of the e-Book. Teacher and student responses averaged 90% and 85.5%, respectively, indicating that the intertextual-based e-book on buffer solutions is highly suitable as a self-learning resource to improve conceptual understanding.</p>

Exploring the role of disciplinary knowledge in students’ covariational reasoning during graphical interpretation

BackgroundThis study investigates undergraduate STEM students’ interpretation of quantities and quantitative relationships on graphical representations in biology (population growth) and chemistry (titration) contexts. Interviews (n = 15) were conducted to explore the interplay between students’ covariational reasoning skills and their use of disciplinary knowledge to form mental images during graphical interpretation.ResultsOur findings suggest that disciplinary knowledge plays an important role in students’ ability to interpret scientific graphs. Interviews revealed that using disciplinary knowledge to form mental images of represented quantities may enhance students’ covariational reasoning abilities, while lacking it may hinder more sophisticated covariational reasoning. Detailed descriptions of four students representing contrasting cases are analyzed, showing how mental imagery supports richer graphic sense-making.ConclusionsIn the cases examined here, students who have a deep understanding of the disciplinary concepts behind the graphs are better able to make accurate interpretations and predictions. These findings have implications for science education, as they suggest instructors should focus on helping students to develop a deep understanding of disciplinary knowledge in order to improve their ability to interpret scientific graphs.

Multi-representation Analysis of General Chemistry Books on Chemical Bonding Subject

This research is a preliminary study that focuses on multi-representation analysis of the chemical bonding topic in three general chemistry of text books. The analysis was carried out to find out how concepts are explained based on the three levels of chemical representation, especially on ionic and covalent bonding, which are the basis for further research in developing intertextual-based learning strategies. Analysis was carried out at the macroscopic-symbolic and sub microscopic-symbolic levels usingmulti-representation analysis table that was adopted from Gkitzia et al., 201 for three general chemistry books, that are 1) Whitten, K.W, et.al. (2014). Chemistry 10th edition. English: Brooks/Cole Cengage Learning; 2) Chang, R. (2010). Chemistry 10th edition. New York : The McGraw-Hill Companies; and 3) Silberberg, M.S, and Amateis, P.G. (2021). Chemistry: the molecular nature of matter and change 9th edition. New York : The McGraw-Hill Companies. The analysis results show that book: Silberberg, M.S, and Amateis, P.G. (2021). Chemistry: the molecular nature of matter and change 9th edition. New York : The McGraw-Hill Companies has the most complete macroscopic-symbolic and sub microscopic-symbolic representation levels on the four main concept labels of chemical bonding, so that they can become standard reference material in developing chemical representations that will be used in intertextual learning strategies.

CPSign: conformal prediction for cheminformatics modeling

Conformal prediction has seen many applications in pharmaceutical science, being able to calibrate outputs of machine learning models and producing valid prediction intervals. We here present the open source software CPSign that is a complete implementation of conformal prediction for cheminformatics modeling. CPSign implements inductive and transductive conformal prediction for classification and regression, and probabilistic prediction with the Venn-ABERS methodology. The main chemical representation is signatures but other types of descriptors are also supported. The main modeling methodology is support vector machines (SVMs), but additional modeling methods are supported via an extension mechanism, e.g. DeepLearning4J models. We also describe features for visualizing results from conformal models including calibration and efficiency plots, as well as features to publish predictive models as REST services. We compare CPSign against other common cheminformatics modeling approaches including random forest, and a directed message-passing neural network. The results show that CPSign produces robust predictive performance with comparative predictive efficiency, with superior runtime and lower hardware requirements compared to neural network based models. CPSign has been used in several studies and is in production-use in multiple organizations. The ability to work directly with chemical input files, perform descriptor calculation and modeling with SVM in the conformal prediction framework, with a single software package having a low footprint and fast execution time makes CPSign a convenient and yet flexible package for training, deploying, and predicting on chemical data. CPSign can be downloaded from GitHub at https://github.com/arosbio/cpsign.Scientific contribution CPSign provides a single software that allows users to perform data preprocessing, modeling and make predictions directly on chemical structures, using conformal and probabilistic prediction. Building and evaluating new models can be achieved at a high abstraction level, without sacrificing flexibility and predictive performance—showcased with a method evaluation against contemporary modeling approaches, where CPSign performs on par with a state-of-the-art deep learning based model.

Is a more physical representation of aerosol chemistry needed for fog forecasting?

AbstractWith the changing climate, the study of fog formation is essential due to the impact of the complexity of natural and anthropogenic aerosols. The evolution of the droplet size distribution in the presence of different aerosol species remains poorly understood. To make progress towards reducing the uncertainty of fog forecasts, the Eulerian–Lagrangian particle‐based small‐scale model for the diffusional growth of droplets is used to better understand the droplet activation and growth. The small‐scale model simulations are performed using observed data from the Winter Fog Experiment study over Indira Gandhi International Airport, New Delhi. The microphysical properties, such as droplet number concentrations (Nd) and liquid water content (LWC), important for fog simulation, are evaluated to gain more insights. The small‐scale simulations have shown the droplet microphysical properties at different evolutionary stages. The Nd and effective radius change with variations in LWC for different aerosol chemistries (i.e., organics, mix, and inorganic). The calculated visibility at small scale is also shown with the variation of Nd and LWC. This study compared visibility from an existing parametrization with parcel–direct numerical simulation calculation. The hygroscopicity , which is highly related to the activation of aerosols to condensation nuclei, is taken into account to demonstrate the contribution of aerosol chemistry to fog droplet formation. The results highlight that hygroscopicity is essential in the numerical model for fog and visibility prediction as the microphysical properties of fog are regulated by aerosol species.

The Effect of the Discovery Learning Model with a Scientific Approach on Student Representation Ability in the Buffer Solution

This study aims to investigate the impact of the discovery learning model with a scientific approach on students' ability to represent macroscopic, submicroscopic, and symbolic aspects of buffer solutions. This study is quantitative research of the quasi-experimental type, utilizing a one group pretest-posttest design. The sampling technique employed is saturated sampling, where the entire population in the study becomes the research sample, consisting of 44 students from Bala Keselamatan Palu High School. The instrument used to measure the three abilities of student representation employs six essay questions, with each degree of representation assessed by two questions. Prior to usage, the six questions are initially verified by an expert of chemical representation abilities. The data analysis techniques used consist of descriptive data analysis and inferential data analysis. The results of the descriptive data analysis indicate that students' representational abilities have improved after participating in learning using the discovery learning model with a scientific approach. The ability of macroscopic representation has a higher average value and percentage of achievement indicators compared to symbolic and submicroscopic representation abilities. The inferential statistical analysis results indicate that the significant value of the Wilcoxon test is 0.000. The value is smaller than the significance level of 0.05, hence it can be concluded that the discovery learning model with a scientific approach influences students' representation abilities.

Modeling the Impacts of Volatile Chemical Product Emissions on Atmospheric Photochemistry and Ozone Formation in Los Angeles

AbstractThe dominant fraction of anthropogenic volatile organic compound (VOC) emissions shifted from transportation fuels to volatile chemical products (VCP) in Los Angeles (LA) in 2010. This shift in VOC composition raises the question about the importance of VCP emissions for ozone (O3) formation. In this study, O3 chemistry during the CalNex 2010 was modeled using the Master Chemical Mechanism (MCM) version 3.3.1 and a detailed representation of VCP emissions based on measurements combined with inventory estimates. The model calculations indicate that VCP emissions contributed to 23% of the mean daily maximum 8‐hr average O3 (DMA8 O3) during the O3 episodes. The simulated OH reactivity, including the contribution from VCP emissions, aligns with observations. Additionally, this framework was employed using four lumped mechanisms with simplified representations of emissions and chemistry. RACM2‐VCP showed the closest agreement with MCM, with a slight 4% increase in average DMA8 O3 (65 ± 13 ppb), whereas RACM2 (58 ± 13 ppb) and SAPRC07B (59 ± 14 ppb) exhibited slightly lower levels. CB6r2, however, recorded reduced concentrations (37 ± 10 ppb). Although emissions of O3 precursors have declined in LA since 2010, O3 levels have not decreased significantly. Model results ascribed this trend to the rapid reduction in NOX emissions. Moreover, given the impact of COVID‐19, an analysis of 2020 reveals a shift to a NOX‐limited O3 formation regime in LA, thereby diminishing the influence of VCPs. This study provides new insights into the impact of VCP emissions on O3 pollution from an in‐depth photochemical perspective.

Application of Transformers in Cheminformatics.

By accelerating time-consuming processes with high efficiency, computing has become an essential part of many modern chemical pipelines. Machine learning is a class of computing methods that can discover patterns within chemical data and utilize this knowledge for a wide variety of downstream tasks, such as property prediction or substance generation. The complex and diverse chemical space requires complex machine learning architectures with great learning power. Recently, learning models based on transformer architectures have revolutionized multiple domains of machine learning, including natural language processing and computer vision. Naturally, there have been ongoing endeavors in adopting these techniques to the chemical domain, resulting in a surge of publications within a short period. The diversity of chemical structures, use cases, and learning models necessitate a comprehensive summarization of existing works. In this paper, we review recent innovations in adapting transformers to solve learning problems in chemistry. Because chemical data is diverse and complex, we structure our discussion based on chemical representations. Specifically, we highlight the strengths and weaknesses of each representation, the current progress of adapting transformer architectures, and future directions.

Exploring Conditions for Utilizing Representations in Chemistry in an Argument-Based Inquiry Environment: Laboratory Only, Technology Only, or a Combination of Laboratory and Technology

Exploring Conditions for Utilizing Representations in Chemistry in an Argument-Based Inquiry Environment: Laboratory Only, Technology Only, or a Combination of Laboratory and Technology

Development of Electronic Student Worksheet Inquiry-Based with Multiple Representations of Chemistry on Reaction Rate Material to Practice Students Critical Thinking Skills

Development of Electronic Student Worksheet Inquiry-Based with Multiple Representations of Chemistry on Reaction Rate Material to Practice Students Critical Thinking Skills

Quantifying the Hardness of Bioactivity Prediction Tasks for Transfer Learning.

Today, machine learning methods are widely employed in drug discovery. However, the chronic lack of data continues to hamper their further development, validation, and application. Several modern strategies aim to mitigate the challenges associated with data scarcity by learning from data on related tasks. These knowledge-sharing approaches encompass transfer learning, multitask learning, and meta-learning. A key question remaining to be answered for these approaches is about the extent to which their performance can benefit from the relatedness of available source (training) tasks; in other words, how difficult ("hard") a test task is to a model, given the available source tasks. This study introduces a new method for quantifying and predicting the hardness of a bioactivity prediction task based on its relation to the available training tasks. The approach involves the generation of protein and chemical representations and the calculation of distances between the bioactivity prediction task and the available training tasks. In the example of meta-learning on the FS-Mol data set, we demonstrate that the proposed task hardness metric is inversely correlated with performance (Pearson's correlation coefficient r = -0.72). The metric will be useful in estimating the task-specific gain in performance that can be achieved through meta-learning.

Neural network kinetics for exploring diffusion multiplicity and chemical ordering in compositionally complex materials

Diffusion involving atom transport from one location to another governs many important processes and behaviors such as precipitation and phase nucleation. The inherent chemical complexity in compositionally complex materials poses challenges for modeling atomic diffusion and the resulting formation of chemically ordered structures. Here, we introduce a neural network kinetics (NNK) scheme that predicts and simulates diffusion-induced chemical and structural evolution in complex concentrated chemical environments. The framework is grounded on efficient on-lattice structure and chemistry representation combined with artificial neural networks, enabling precise prediction of all path-dependent migration barriers and individual atom jumps. To demonstrate the method, we study the temperature-dependent local chemical ordering in a refractory NbMoTa alloy and reveal a critical temperature at which the B2 order reaches a maximum. The atomic jump randomness map exhibits the highest diffusion heterogeneity (multiplicity) in the vicinity of this characteristic temperature, which is closely related to chemical ordering and B2 structure formation. The scalable NNK framework provides a promising new avenue to exploring diffusion-related properties in the vast compositional space within which extraordinary properties are hidden.

A better representation of VOC chemistry in WRF-Chem and its impact on ozone over Los Angeles.

The declining trend in vehicle emissions has underscored the growing significance of Volatile Organic Compound (VOC) emissions from Volatile Chemical Products (VCP). However, accurately representing VOC chemistry in simplified chemical mechanisms remains challenging due to its chemical complexity including speciation and reactivity. Previous studies have predominantly focused on VOCs from fossil fuel sources, leading to an underrepresentation of VOC chemistry from VCP sources. We developed an integrated chemical mechanism, RACM2B-VCP, that is compatible with WRF-Chem and is aimed to enhance the representation of VOC chemistry, particularly from VCP sources, within the present urban environment. Evaluation against the Air Quality System (AQS) network data demonstrates that our model configured with RACM2B-VCP reproduces both the magnitude and spatial variability of O3 as well as PM2.5 in Los Angeles. Furthermore, evaluation against comprehensive measurements of O3 and PM2.5 precursors from the Reevaluating the Chemistry of Air Pollutants in California (RECAP-CA) airborne campaign and the Southwest Urban NO x and VOC Experiment (SUNVEx) ground site and mobile laboratory campaign, confirm the model's accuracy in representing NOx and many VOCs and highlight remaining biases. Although there exists an underprediction in the total VOC reactivity of observed VOC species, our model with RACM2B-VCP exhibits good agreement for VOC markers emitted from different sectors, including biogenic, fossil fuel, and VCP sources. Through sensitivity analyses, we probe the contributions of VCP and fossil fuel emissions to total VOC reactivity and O3. Our results reveal that 52% of the VOC reactivity and 35% of the local enhancement of MDA8 O3 arise from anthropogenic VOC emissions in Los Angeles. Significantly, over 50% of this anthropogenic fraction of either VOC reactivity or O3 is attributed to VCP emissions. The RACM2B-VCP mechanism created, described, and evaluated in this work is ideally suited for accurately representing ozone for the right reasons in the present urban environment where mobile, biogenic, and VCP VOCs are all important contributors to ozone formation.

Enhancing Students' Conceptual Understanding of Chemistry in a SiMaYang Learning Environment

Students’ conceptual understanding of chemistry needs to be supported by the skill to think at three levels of representation. Many students have difficulty understanding chemistry concepts because of it. Chemistry teaching needs to provide a learning environment that involves chemical representations to facilitate students’ interconnection skills of three levels of representation and ultimately improve students’ conceptual understanding. SiMaYang learning was designed by integrating chemical representation during the learning process. This study aimed to enhance students’ conceptual understanding of chemistry through the implementation of SiMaYang learning. A descriptive quantitative research method with one group pretest-posttest design was used in this study. Students’ conceptual understanding was measured using three sets of essay tests containing macroscopic, symbolic, and submicroscopic tests for the topic of the buffer concept. The research data were analyzed descriptively and statistically using N-gain and paired sample t-tests. The results of the paired sample t-test showed the sig value. 0.000 <0.05, which means SiMaYang learning affects increasing students’ conceptual understanding of chemistry. The finding of this study informed the students’ conceptual understanding of chemistry increased in the medium category (N-gain = 0.62). SiMaYang learning, which was designed to involve three levels of representation, can increase students’ conceptual understanding of chemistry. Keywords: chemistry, conceptual understanding ,SiMaYang learning environment