We offer a variety of services focused on bioinformatics-based integration of metabolomics, proteomics and transcriptomics. Our bioinformatics consultants can assist PIs with Gene Set Enrichment Analysis (GSEA), also assist with biomarker discovery projects by building predictive/classification models, depending on the interests of the study, and with using machine learning techniques such as Classification and Regression Trees (CART), Multivariate Adaptive Regression Splines (MARS), Random Forests (RF), and other bootstrapping/ensemble methods.
If you have any questions about these services or about the Informatics Core, please reach out to firstname.lastname@example.org, and please feel free to submit a request for services at https://utmb.us/2ae.
To realize the full potential of the data generated by omics facilities, bioinformatics-based integration of metabolomics, proteomics, and transcriptomics is essential. Molecular profiles derived from omics data can be subjected to analyses, which can then be augmented with public data. This integration is conducted by either using pathway enrichment tools or modeling as a central step.
For studies involving complex integration of heterogeneous data, we offer Monte-Carlo based approach to assess the statistical significance of the output of such analysis.
For pathway enrichment tools, the significant results from omics studies (namely a list of genes, proteins, and metabolites that are significantly differentially expressed at the 0.05 FDR level of significance) are used as inputs into IPA software and GSEA. Our goal is to build biological context around these results, allowing us to identify the interlocking pathways, co-regulated sub-networks, and critical nodes that map to study-specific pathways and networks.
For biomarker discovery, we can use machine learning techniques such as Classification and Regression Trees (CART), Multivariate Adaptive Regression Splines (MARS), Random Forests (RF), and other bootstrapping/ensemble methods to create predictive models. These models can then be used to identify characteristics important to the study at the molecular and physiological levels.
The results of CART, MARS, and/or RF will be a list of genes, proteins, metabolites, clinical and/or demographic characteristics. From the list of variables in the models, we can use IPA to analyze those genes, proteins, and/or metabolites to identify relevant pathways of interest associated with the important genes, proteins, and/or metabolites.