Development of a data processing pipeline for multiplatform phenotyping experiments

Details

Supervisors

Draye, Xavier

Faculty

Faculté des bioingénieurs

Degree label

Master [120] : bioingénieur en sciences agronomiques, à finalité spécialisée

Abstract

The thesis aims to establish a phenotypic data analysis pipeline, enabling the creation of statistical models. An initial literature review on maize phenotyping in general highlights the benefits of various phenotyping infrastructures, as well as the challenges faced by the phenotyping community in the field daily. Next, the work outlines the project objectives. Within the context of the challenges encountered by phenotyping researchers, it explores how the pipeline can assist in addressing these issues and in what ways it provides new solutions compared to existing applications. Following this, the next section is dedicated to the methodology employed. To implement the pipeline, the R programming language was utilized, specifically the statgenHTP package enabling statistical modeling for analyzing data coming from high throughput phenotyping platform experiments. The results of the different analyses are presented. For each analysis, graphs illustrate the evolution of various parameters such as biomass or plant height. At the conclusion of the results, a critical analysis is conducted to assess the relevance of our pipeline compared to phenotyping standards. Lastly, a discussion on the overall project, primarily focusing on statistical modeling, wraps up the report.