Filtern
Erscheinungsjahr
- 2024 (21) (entfernen)
Dokumenttyp
- Wissenschaftlicher Artikel (9)
- Konferenzveröffentlichung (6)
- Preprint (3)
- Sonstiges (2)
- Lehrmaterial (1)
Schlagworte
- DECIMER (2)
- Deep Learning (2)
- Dissipative Particle Dynamics (2)
- OCSR, Optical Chemical Structure Recognition (2)
- Transformer (2)
- Abusive Supervision (1)
- AlphaFold, ColabFold, PyMOL (1)
- Amylase, Enzymcharakterisierung (1)
- Augmented Multiphase Rail Launcher (1)
- Bone Morphogenetic Protein, BMP, BMP2 (1)
Advancements in hand-drawn chemical structure recognition through an enhanced DECIMER architecture
(2024)
Accurate recognition of hand-drawn chemical structures is crucial for digitising hand-written chemical information in traditional laboratory notebooks or facilitating stylus-based structure entry on tablets or smartphones. However, the inherent variability in hand-drawn structures poses challenges for existing Optical Chemical Structure Recognition (OCSR) software. To address this, we present an enhanced Deep lEarning for Chemical ImagE Recognition (DECIMER) architecture that leverages a combination of Convolutional Neural Networks (CNNs) and Transformers to improve the recognition of hand-drawn chemical structures. The model incorporates an EfficientNetV2 CNN encoder that extracts features from hand-drawn images, followed by a Transformer decoder that converts the extracted features into Simplified Molecular Input Line Entry System (SMILES) strings. Our models were trained using synthetic hand-drawn images generated by RanDepict, a tool for depicting chemical structures with different style elements. A benchmark was performed using a real-world dataset of hand-drawn chemical structures to evaluate the model's performance. The results indicate that our improved DECIMER architecture exhibits a significantly enhanced recognition accuracy compared to other approaches.
Advancements in Hand-Drawn Chemical Structure Recognition through an Enhanced DECIMER Architecture
(2024)
Accurate recognition of hand-drawn chemical structures is crucial for digitising hand-written chemical information found in traditional laboratory notebooks or for facilitating stylus-based structure entry on tablets or smartphones. However, the inherent variability in hand-drawn structures poses challenges for existing Optical Chemical Structure Recognition (OCSR) software. To address this, we present an enhanced Deep lEarning for Chemical ImagE Recognition (DECIMER) architecture that leverages a combination of Convolutional Neural Networks (CNNs) and Transformers to improve the recognition of hand-drawn chemical structures. The model incorporates an EfficientNetV2 CNN encoder that extracts features from hand-drawn images, followed by a Transformer decoder that converts the extracted features into Simplified Molecular Input Line Entry System (SMILES) strings. Our models were trained using synthetic hand-drawn images generated by RanDepict, a tool for depicting chemical structures with different style elements. To evaluate the model's performance, a benchmark was performed using a real-world dataset of hand-drawn chemical structures. The results indicate that our improved DECIMER architecture exhibits a significantly enhanced recognition accuracy compared to other approaches.
Inspired by the super-human performance of deep learning models in playing the game of Go after being presented with virtually unlimited training data, we looked into areas in chemistry where similar situations could be achieved. Encountering large amounts of training data in chemistry is still rare, so we turned to two areas where realistic training data can be fabricated in large quantities, namely a) the recognition of machine-readable structures from images of chemical diagrams and b) the conversion of IUPAC(-like) names into structures and vice versa. In this talk, we outline the challenges, technical implementation and results of this study.
Optical Chemical Structure Recognition (OCSR): Vast amounts of chemical information remain hidden in the primary literature and have yet to be curated into open-access databases. To automate the process of extracting chemical structures from scientific papers, we developed the DECIMER.ai project. This open-source platform provides an integrated solution for identifying, segmenting, and recognising chemical structure depictions in scientific literature. DECIMER.ai comprises three main components: DECIMER-Segmentation, which utilises a Mask-RCNN model to detect and segment images of chemical structure depictions; DECIMER-Image Classifier EfficientNet-based classification model identifies which images contain chemical structures and DECIMER-Image Transformer which acts as an OCSR engine which combines an encoder-decoder model to convert the segmented chemical structure images into machine-readable formats, like the SMILES string.
DECIMER.ai is data-driven, relying solely on the training data to make accurate predictions without hand-coded rules or assumptions. The latest model was trained with 127 million structures and 483 million depictions (4 different per structure) on Google TPU-V4 VMs
Name to Structure Conversion: The conversion of structures to IUPAC(-like) or systematic names has been solved algorithmically or rule-based in satisfying ways. This fact, on the other side, provided us with an opportunity to generate a name-structure training pair at a very large scale to train a proof-of-concept transformer network and evaluate its performance.
In this work, the largest model was trained using almost one billion SMILES strings. The Lexichem software utility from OpenEye was employed to generate the IUPAC names used in the training process. STOUT V2 was trained on Google TPU-V4 VMs. The model's accuracy was validated through one-to-one string matching, BLEU scores, and Tanimoto similarity calculations. To further verify the model's reliability, every IUPAC name generated by STOUT V2 was analysed for accuracy and retranslated using OPSIN, a widely used open-source software for converting IUPAC names to SMILES. This additional validation step confirmed the high fidelity of STOUT V2's translations.
An Augmented Multiphase Rail Launcher With a Modular Design: Extended Setup and Muzzle Fed Operation
(2024)
An automated pipeline for comprehensive calculation of intermolecular interaction energies based on molecular force-fields using the Tinker molecular modelling package is presented. Starting with non-optimized chemically intuitive monomer structures, the pipeline allows the approximation of global minimum energy monomers and dimers, configuration sampling for various monomer-monomer distances, estimation of coordination numbers by molecular dynamics simulations, and the evaluation of differential pair interaction energies. The latter are used to derive Flory-Huggins parameters and isotropic particle-particle repulsions for Dissipative Particle Dynamics (DPD). The computational results for force fields MM3, MMFF94, OPLSAA and AMOEBA09 are analyzed with Density Functional Theory (DFT) calculations and DPD simulations for a mixture of the non-ionic polyoxyethylene alkyl ether surfactant C10E4 with water to demonstrate the usefulness of the approach.
An automated pipeline for comprehensive calculation of intermolecular interaction energies based on molecular force-fields using the Tinker molecular modelling package is presented. Starting with non-optimized chemically intuitive monomer structures, the pipeline allows the approximation of global minimum energy monomers and dimers, configuration sampling for various monomer-monomer distances, estimation of coordination numbers by molecular dynamics simulations, and the evaluation of differential pair interaction energies. The latter are used to derive Flory-Huggins parameters and isotropic particle-particle repulsions for Dissipative Particle Dynamics (DPD). The computational results for force fields MM3, MMFF94, OPLS-AA and AMOEBA09 are analyzed with Density Functional Theory (DFT) calculations and DPD simulations for a mixture of the non-ionic polyoxyethylene alkyl ether surfactant C10E4 with water to demonstrate the usefulness of the approach.
Einleitung und Fragestellung:
Abusive Supervision wird mit willentlicher Leistungszurückhaltung, verringerter Motivation, erhöhtem Stresserleben, psychosomatischen Beschwerden und Burnout bei Mitarbeitenden assoziiert. Angesichts der hohen Prävalenz destruktiver Führung bleibt bislang die Frage offen, welche
protektiven Ressourcen die genannten Zusammenhänge abpuffern.
Theoretischer Hintergrund:
Abusive Supervision bezieht sich auf das Ausmaß der feindseligen verbalen und nonverbalen Verhaltensweisen einer Führungskraft. Basierend auf dem Anforderungs- Ressourcen- Modell gehen wir davon aus, dass sich personale Ressourcen, die Mitarbeitende in der arbeitsfreien Zeit aufbauen, positiv auf den negativen Effekt zwischen destruktiver Führung und Mitarbeitergesundheit auswirken. Wir fokussieren hier die generalisierte Selbstwirksamkeitserwartung, die sich im Sinne der sozialkognitiven Theorie und zahlreichen empirischen Befunden als gesundheitsrelevante Ressource im
Umgang mit domänenübergreifenden Belastungen herausgestellt hat. Diese sollte durch Bewältigungserfahrung in der arbeitsfreien Zeit gefördert werden. Bewältigungserfahrung in der Freizeit bedeutet die Gelegenheit des Erlebens von Kompetenz und Fachwissen.
Methode:
Die Moderatoranalyse wurde im Rahmen einer Querschnittsbefragung einer anfallenden Stichprobe mit N = 305 Personen getestet. Die Variablen wurden mit der Abusive Supervision Scale (Tepper, 2000), dem REQ (Sonnentag & Fritz, 2007), und der Subskala emotionale Erschöpfung des MBI (Büssing & Perrar, 1992) gemessen.
Ergebnisse:
In dieser Studie zeigen „Mastery Experiences“ einen hypothesenkonformen Puffereffekt, nicht jedoch die anderen Erholungsstrategien, die auch mit getestet wurden. Es zeigt sich also die Tendenz, dass sich Mitarbeitende durch das Erlernen neuer Kompetenzen und den Aufbau von Selbstwirksamkeit vor den gesundheitsschädlichen Auswirkungen destruktiver Führung schützen können. Das
Korrelationsmuster deutet aber vrmtl. auch problematische Aspekte dieser Erholungsstrategie an.
Diskussion:
Limitierend muss erwähnt werden, dass wir die vermutete vermittelnde Variable Selbstwirksamkeit nicht explizit gemessen haben, und dass zukünftige Untersuchungen den Effekt in Form einer mediierten Moderation replizieren müssen.
ChatGPT ist ein leistungsstarker Chatbot, der nach Eingabe konkreter Aufforderungen maßgeschneiderte Texte erstellt und Entwickler beim Programmieren unterstützen kann. Dazu bildet das GPT-Modell, ein „Large Language Model“ (LLM), Muster auf ein statistisches Modell ab, die dem Nutzer eine Antwort auf eine Frage generieren. Durch die große mediale Aufmerksamkeit mit der ChatGPT eingeführt wurde haben eine Vielzahl von Nutzern die potenziellen Chancen dieser Technologie kennengelernt. Jedoch birgt ChatGPT auch eine Reihe von Risiken.
In diesem Artikel werden sowohl die Chancen als auch die Risiken von ChatGPT umfassend insbesondere im Bereich Cyber-Sicherheit betrachtet.
As a rule, an experiment carried out at school or in undergraduate study
courses is rather simple and not very informative. However, when the experiments
are to be performed using modern methods, they are often abstract and
difficult to understand. Here, we describe a quick and simple experiment,
namely the enzymatic characterization of ptyalin (human salivary amylase)
using a starch degradation assay. With the experimental setup presented here,
enzyme parameters, such as pH optimum, temperature optimum, chloride
dependence, and sensitivity to certain chemicals can be easily determined. This
experiment can serve as a good model for enzyme characterization in general,
as modern methods usually follow the same principle: determination of the
activity of the enzyme under different conditions. As different alleles occur in
humans, a random selection of test subjects will be quite different with regard
to ptyalin activities. Therefore, when the students measure their own ptyalin
activity, significant differences will emerge, and this will give them an idea of
the genetic diversity in human populations. The evaluation has shown that the
pupils have gained a solid understanding of the topic through this experiment.