More than 30 years after the nuclear power plant accident in Chornobyl, wildlife in the Chornobyl Exclusion Zone (CEZ) is still chronically exposed to low doses of ionizing radiation. In 2018, populations of the Eastern tree frog Hyla orientalis were sampled inside the CEZ across a gradient of radiocontamination and in nearby control sites. Our research project aims at developing methods for multi-omics data integration to advance the understanding of the consequences of chronic exposure to low-dose radiation. The first axis of this study focuses on the exploration of transcriptomic data in relation to dosimetry. The analysis of RNA-seq data could reveal radiation-specific molecular signatures; however, this can only be accomplished if confounding factors are properly accounted for. Here, we compare batch effect removal strategies using the transcriptomic data obtained from 87 tree frogs. We highlight the strengths and weaknesses of the following methods: the ComBat-seq method dedicated for RNA-seq count data, linear regression on the batch with residuals extraction, and an integrated approach with mixOmics MINT. The geographical site of origin of the frogs is a confounding factor for the analysis, and its complete confoundedness with the dose makes its handling challenging. Indeed, the different methods we used were brought to their limits: correcting or accounting for the batch effect cuts out the biological information related to the dose. The site factor encompasses hidden information, including the local environment and available food sources, the day of capture, and the genetic diversity between the populations of tree frogs. In a second approach, we performed hierarchical clustering on the genetic distance matrix to summarize the genetic diversity into a batch factor. Using the methods previously mentioned, we were able to integrate genetic diversity into the analysis and bring to light the effect of the ionizing radiation dose in the transcriptomic data.