Metabolic state prediction at Single-cell resolution

Abstract

La diversitat en el metabolisme condueix a poder observar diferents caracteristiques cel・lulars dins un mateix teixit. Aquest concepte, anomenat heterogeneitat tissular, es fonamental en els sistemes biologics. En el context del cancer, una de les principals causes de mort arreu del mon, l'heterogeneitat tissular te un paper clau. L'heterogeneitat dels tumors contribueix a la complexitat de la progressio del cancer, la metastasi i la resistencia als tractaments actuals. Aquest treball final de grau explora el desenvolupament i l'aplicacio d'eines computacionals relacionades amb la transcriptomica i la metabolomica en l’estudi de l'heterogeneitat de fenotips. La part central del projecte va implicar la construccio d'una eina computacional per integrar dades scRNA-seq amb models metabolics, quelcom que no s'havia fet anteriorment. Mitjancant l'analisi de dinamiques metaboliques a escala de cel・lula individual es poden obtenir noves interpretacions biologiques. Aquest enfocament ofereix una nova perspectiva per a estudiar el fenotip cel・lular, potencialment aplicable en biotecnologia i biomedicina. El procediment desenvolupat es va provar en el conjunt de dades PBMC 3k de 10x Genomics i la reconstruccio metabolica humana Recon3D de la base de dades VMH. Els resultats principals del projecte inclouen la generacio de models metabolics a partir de dades de scRNA-seq, la identificació de diferents grups metabolics i la visualitzacio de canvis metabolics de diferents estats cel・lulars. La nova eina proporciona un recurs per a la comunitat cientifica que permet l'analisi de la dinàmica metabolica a escala de cel・lula individual i ofereix informacio sobre el paper del metabolisme en diferents contextos. Els resultats del projecte mostren com la combinacio de la transcriptomica amb la metabolomica pot revelar noves troballes biologiques, obrint possibilitats per a mes investigacions en aquest camp. Aixo fomenta la innovacio i el desenvolupament de tecnologia d'acces obert, proporcionant eines per a l’estudi del funcionament de malalties, d'acord amb els Objectius de Desenvolupament Sostenible de "salut i benestar" i "industria, innovacio i infraestructura".

Diversity in metabolism leads to variation in cellular characteristics observed within the same tissue. This concept, called tissue heterogeneity, is a pivotal aspect of biological systems. In the context of cancer, found in the top causes of death worldwide, tissue heterogeneity plays a critical role. Heterogeneity within tumors contributes to the complexity of cancer progression, metastasis, and treatment resistance. This bachelor’s thesis explores the development and application of computational tools related to transcriptomics and metabolomics to study phenotype heterogeneity. The project’s core involved constructing a computational pipeline for integrating scRNA-seq data with metabolic models, which has not been done before. By analyzing metabolic dynamics at the singlecell level, novel biological interpretations can be obtained. This approach offers a new perspective to studying cellular phenotype, potentially applicable in biotechnology and biomedicine. The framework was tested using the PBMC 3k dataset from 10x Genomics and the human metabolic network reconstruction Recon3D from the VMH database. The primary outcomes of the project, including the generation of metabolic models from single-cell data, the identification of distinct metabolic clusters, and the visualization of metabolic changes across different cell states, have practical implications. This tool provides a resource for the scientific community, enabling analysis of metabolic dynamics at the single-cell level and offering insights into context-specific cellular metabolism's role. The project’s results show how combining transcriptomics with metabolomics can reveal new biological insights, opening up possibilities for further research in this field. This thesis fosters innovation and the development of open-access technology, providing tools for studying health and disease mechanisms, in line with the Sustainable Development Goals of “health and wellness” and “industry, innovation, and infrastructure”.