Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

La recherche d’information biologiquement pertinente à partir de données « omics » est un défi majeur dans les analyses d’expression à l’échelle du génome. Les données « omics » ne génèrent pas toutes la même quantité de variables. Cependant, les algorithmes de détection de voies de signalisation sont utilisés uniformément. La question de la relevance des voies de signalisation détectées avec des jeux de données moins denses en information peut donc se poser. Dans cette étude, nous évaluons comment la variation de la quantité de data background dans un jeu de données transcriptomiques influe sur les résultats générés par GSEA, l’algorithme le plus communément utilisé par la communauté scientifique pour la recherche de voies de signalisation biologiques.

RNA-seq, NGS, GSEA, DESeq2, bioinformatics, biostatistics --- Sciences de la santé humaine > Médecine de laboratoire & technologie médicale

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Systemic AutoInflammatory Disorders or SAIDs constitute rare diseases where the innate immune response is dysregulated. Though there are genes mutations that are associated with monogenic SAIDs, polygenic ones have higher incidence and are more difficult to diagnose. Ambiotis, a CRO based in Toulouse (FR) and with an expertise in the resolution of inflammation hypothesize SAIDs patients have a dysregulated resolution of inflammation. In the context of the Immunome consortium for AutoInflammatory Disorders or ImmunAID, a European Union funded project, a cohort of SAIDs patients was created throughout Europe to collect biological sample. These sample would be analysed by multiple omics techniques to acquire the immunome of SAIDs as a group and as the different diseases. Ambiotis, part of the ImmunAID consortium, quantified Specialized Pro-resolution Mediators (SPM) using UHPLC-MS technique. This master’s thesis goal was to design a pipeline and R library to process the lipidomic data, run multiple pairwise comparison and display results in different graphs format. Our pipeline was able to identify protectin D1 having higher concentration in some SAIDs when compared to negative controls.
These results were obtained on the quantified data available; the next step will be to use the entire data set. In the end, we managed to build a pipeline that correct the batch effect, normalize the data and run multiple pairwise comparisons.

SAIDs, autoinflammation, data analysis, computational biology --- Sciences de la santé humaine > Médecine de laboratoire & technologie médicale