This is a study concerning the use of Machine Learning (ML) techniques to ascertain the impacts of particle ionizing radiation (IR) on cell survival and DNA damage. Current empirical models do not always take into account intrinsic complexities and stochastic effects of the interactions of IR and cell populations. Furthermore, these models often lack in biophysical interpretations of the irradiation outcomes. The linear quadratic (LQ) model is a common way to associate the biological response of a cell population with the radiation dose. The parameters of the LQ model are used to extrapolate the relation between the dosage and the survival fraction of a cell population. The goal was to create a ML-based model that predicts the $\alpha$ and $\beta$ parameters of the well known and established LQ model, along with the key metrics of DNA damage induction. The main target of this effort was, on the one hand, the development of a computational framework that will be able to assess key radiobiophysical quantities, and on the other hand, to provide meaningful interpretations of the outputs. Based on our results, as some metrics of the adaptability and training efficiency, our ML models exhibited 0.18 median error (relative root mean squared error (RRMSE)) in the prediction of the $\alpha$ parameter and errors of less than 0.01 for various DNA damage quantities; the prediction for $\beta$ exhibited a rather large error of 0.75. Our study is based on experimental data from a publicly available dataset of irradiation studies. All types of complex DNA damage (all clusters), and the number of double-stranded breaks (DSBs), which are widely accepted to be closely related to cell survival and the detrimental biological effects of IR, were calculated using the fast Monte Carlo Damage Simulation software (MCDS). We critically discussed the varying importance of physical parameters such as charge and linear energy transfer (LET); we also discussed the uncertainties of our predictions and future directions, and the dynamics of our approach.

Based on the need for radiobiological databases, in this work, we mined experimental ionizing radiation data of human cells treated with X-rays, γ-rays, carbon ions, protons and α-particles, by manually searching the relevant literature in PubMed from 1980 until 2024. In order to calculate normal and tumor cell survival α and β coefficients of the linear quadratic (LQ) established model, as well as the initial values of the double-strand breaks (DSBs) in DNA, we used WebPlotDigitizer and Python programming language. We also produced complex DNA damage results through the fast Monte Carlo code MCDS in order to complete any missing data. The calculated α/β values are in good agreement with those valued reported in the literature, where α shows a relatively good association with linear energy transfer (LET), but not β. In general, a positive correlation between DSBs and LET was observed as far as the experimental values are concerned. Furthermore, we developed a biophysical prediction model by using machine learning, which showed a good performance for α, while it underscored LET as the most important feature for its prediction. In this study, we designed and developed the novel radiobiological ‘RadPhysBio’ database for the prediction of irradiated cell survival (α and β coefficients of the LQ model). The incorporation of machine learning and repair models increases the applicability of our results and the spectrum of potential users.

Pancreatic cancer is projected to become the second leading cause of cancer-related deaths by 2030, with its mortality continuing to rise, unlike other common cancers such as breast or colorectal. Late-stage diagnosis, often accompanied by metastatic dissemination, drastically impairs patient survival and underscores the urgent need for improved biomarkers to guide therapeutic strategies. While molecular signatures have been proposed to stratify pancreatic cancer patients, their ability to predict outcomes remains limited. In this study, we applied established molecular signatures to our in-house transcriptomic data from a cohort of pancreatic cancer patients. We took advantage of published datasets to construct comprehensive atlases of cells present in primary and metastatic pancreatic cancers. The atlas of metastasis samples, representative of routinely harvested patient biopsies, revealed that monocyte/macrophage signatures provided superior discriminatory power compared to existing molecular classifications. Notably, the abundance of TREM2-expressing macrophages emerged as a significant parameter for stratifying patients. Our findings position TREM2⁺ macrophages as a promising biomarker for pancreatic cancer, with potential to enhance patient stratification and inform the development of targeted therapies. This work highlights the critical role of tumor-associated macrophages in pancreatic cancer progression and lays the groundwork for further functional and translational studies.

Analyse génétique et transcriptomique de l'adénocarcinome canalaire pancréatique Les monocytes et les macrophages comme outil d'analyse des données moléculaires des patients: Cette thèse de doctorat porte sur la caractérisation moléculaire de patients souffrant d'un adénocarcinome canalaire pancréatique (PDAC), une forme particulièrement agressive et mortelle de cancer.Cette recherche étudie le microenvironnement tumoral du PDAC, en mettant l'accent sur la compréhension du rôle des cellules immunitaires, notamment les macrophages, dans la biologie des tumeurs.Nous avons combiné différentes approches, notamment le profilage transcriptomique et mutationnel ainsi que la transcriptomique spatiale.Nous avons observé qu'une cohorte locale de patients atteints de PDAC pouvait être classée en deux groupes distincts selon leur profil transcriptomique.Cette dichotomie a été caractérisée en comparant notre cohorte à celles disponibles dans la littérature, révélant que les deux groupes peuvent être distingués en fonction de l'abondance des signatures monocytes/macrophages, notamment des macrophages TREM2 extsuperscript{+}.Par la suite, nous avons analysé les mutations génétiques de ces deux groupes et avons observé des différences entre les deux groupes.Enfin, nous avons utilisé la transcriptomique spatiale pour examiner les cellules tumorales et les macrophages dans des échantillons appariés rares de PDAC primaires et des métastases. Nos données ont mis en évidence des variations entre les sites primaires et métastatiques de patients, qui peuvent éclairer des approches de traitement adaptées.Dans l'ensemble, nos résultats soulignent l'importance de l'analyse à résolution spatiale et le potentiel des thérapies ciblées dans les traitements du PDAC.