Datasets Construction and Development of QSAR Models for Predicting Micronucleus In Vitro and In Vivo Assay Outcomes.

In silico (quantitative) structure-activity relationship modeling is an approach that provides a fast and cost-effective alternative to assess the genotoxic potential of chemicals. However, one of the limiting factors for model development is the availability of consolidated experimental datasets. In the present study, we collected experimental data on micronuclei in vitro and in vivo, utilizing databases and conducting a PubMed search, aided by text mining using the BioBERT large language model. Chemotype enrichment analysis on the updated datasets was performed to identify enriched substructures. Additionally, chemotypes common for both endpoints were found. Five machine learning models in combination with molecular descriptors, twelve fingerprints and two data balancing techniques were applied to construct individual models. The best-performing individual models were selected for the ensemble construction. The curated final dataset consists of 981 chemicals for micronuclei in vitro and 1309 for mouse micronuclei in vivo, respectively. Out of 18 chemotypes enriched in micronuclei in vitro, only 7 were found to be relevant for in vivo prediction. The ensemble model exhibited high accuracy and sensitivity when applied to an external test set of in vitro data. A good balanced predictive performance was also achieved for the micronucleus in vivo endpoint.

Improving VAE based molecular representations for compound property prediction.

Collecting labeled data for many important tasks in chemoinformatics is time consuming and requires expensive experiments. In recent years, machine learning has been used to learn rich representations of molecules using large scale unlabeled molecular datasets and transfer the knowledge to solve the more challenging tasks with limited datasets. Variational autoencoders are one of the tools that have been proposed to perform the transfer for both chemical property prediction and molecular generation tasks. In this work we propose a simple method to improve chemical property prediction performance of machine learning models by incorporating additional information on correlated molecular descriptors in the representations learned by variational autoencoders. We verify the method on three property prediction tasks. We explore the impact of the number of incorporated descriptors, correlation between the descriptors and the target properties, sizes of the datasets etc. Finally, we show the relation between the performance of property prediction models and the distance between property prediction dataset and the larger unlabeled dataset in the representation space.

Low-Energy Laser-Driven Ultrashort Pulsed Electron Beam Irradiation-Induced Immune Response in Rats.

The development of new laser-driven electron linear accelerators, providing unique ultrashort pulsed electron beams (UPEBs) with low repetition rates, opens new opportunities for radiotherapy and new fronts for radiobiological research in general. Considering the growing interest in the application of UPEBs in radiation biology and medicine, the aim of this study was to reveal the changes in immune system in response to low-energy laser-driven UPEB whole-body irradiation in rodents. Forty male albino Wistar rats were exposed to laser-driven UPEB irradiation, after which different immunological parameters were studied on the 1st, 3rd, 7th, 14th, and 28th day after irradiation. According to the results, this type of irradiation induces alterations in the rat immune system, particularly by increasing the production of pro- and anti-inflammatory cytokines and elevating the DNA damage rate. Moreover, such an immune response reaches its maximal levels on the third day after laser-driven UPEB whole-body irradiation, showing partial recovery on subsequent days with a total recovery on the 28th day. The results of this study provide valuable insight into the effect of laser-driven UPEB whole-body irradiation on the immune system of the animals and support further animal experiments on the role of this novel type of irradiation.

Laser-Driven Ultrashort Pulsed Electron Beam Radiation at Doses of 0.5 and 1.0 Gy Induces Apoptosis in Human Fibroblasts.

Rapidly evolving laser technologies have led to the development of laser-generated particle accelerators as an alternative to conventional facilities. However, the radiobiological characteristics need to be determined to enhance their applications in biology and medicine. In this study, the radiobiological effects of ultrashort pulsed electron beam (UPEB) and X-ray radiation in human lung fibroblasts (MRC-5 cell line) exposed to doses of 0.1, 0.5, and 1 Gy are compared. The changes of γH2AX foci number as a marker of DNA double-strand breaks (DSBs) were analyzed. In addition, the micronuclei induction and cell death via apoptosis were studied. We found that the biological action of UPEB-radiation compared to X-rays was characterized by significantly slower γH2AX foci elimination (with a dose of 1 Gy) and strong apoptosis induction (with doses of 0.5 and 1.0 Gy), accompanied by a slight increase in micronuclei formation (dose of 1 Gy). Our data suggest that UPEB radiation produces more complex DNA damage than X-ray radiation, leading to cell death rather than cytogenetic disturbance.