Integrative multi-omics analysis provides a more holistic approach for studying the mechanism of diseases and identifying putative biomarkers that are not easily achieved by single -omics analysis. Rett syndrome (RTT) is a neurodevelopmental disorder caused by congenital MECP2 gene mutations and primarily affect females. This study performed transcriptomics, proteomics, phosphoproteomics and metabolomics analyses of blood samples from patients with classical RTT (n=9) and age- and sex-matched controls (n=8). Each -omics dataset was analysed individually and then the multi-omics datasets were analysed collectively using both supervised (MixOmics) and unsupervised [Multi‐Omics Factor Analysis (MOFA2)] machine learning algorithms. To minimise batch effects from each data set, the remove unwanted variation (RUVIII-C) tool was used. Differential abundance and pathway enrichment analysis of the transcriptomic and proteomic data sets showed dysregulation of mRNA processing, mitochondrial function, and ribosome. Changes in amino acid and lipid pathways were observed in the metabolomics datasets. Analysis of phosphoproteomics data highlighted putative dysregulation of upstream kinases involved in regulating cell growth, differentiation and inflammatory response. MixOmics and MOFA2 analyses support the results identified above and highlight a small subset of transcripts, proteins, metabolites that were differentially co-expressed in disease versus controls and could be putative biomarkers. To our knowledge, our study is the first to integrate multi-omics data from the blood of patients with RTT, providing mechanistic insights, discovering potential new biomarkers, and identifying possible new therapeutic targets. This methodology may also be useful in other rare genetic diseases.