Allen Brain Atlases and Data

[cwyoo]

Aging, Dementia and Traumatic Brain Injury Study available at: http://aging.brain-map.org/

[cpere117]

Dear all,

I will be performing RNA-Seq analysis upon the Aging, Dementia, and Traumatic Brain Injury (TBI) Dr. Yoo posted in the link above. To begin, I first wanted to clarify what reasoning and my study objectives are in performing this RNA-Seq analysis and some distinguishable variables that we can discern from this database regarding study samples as compared to the GEO database that often can be more ambiguous in regards to defining clinical variables unique to each sample in a particular dataset. This repository provides RNA sequencing datasets from 107 post-mortem brain samples each corresponding to a unique TBI cohort patient's brain, including in total 377 brain tissue samples across brain regions that include a diverse array of brain structures consisting from cortical grey (parietal and temporal) and white matter (parietal) and hippocampus. Some of you may be wondering how I could have 107 patients but have a total of 377 tissue samples. That is a good observation and can be clarified by the fact that researchers in this study used multiple tissue samples for most individual patient brains to measure the endogenous shift in RNA protein expression across an individuals brain regions. In a preliminary analysis using the RNA-Seq tab in the link above you can view the expression of a gene of interest (for me ID3) analyzed and mapped across all tissue samples. Furthermore, you can define your variables of interest when constructing the heat map which in my case was gender, brain region, and dementia disease state. When choosing these parameters we found that the ID family with ID3 most specifically being much more highly expressed in diseased AD Hippocampus regions of Females and Males being more highly expressed in dementia diagnosed patients versus non-Dementia diagnosed patients. Therefore, to further validate this tool's output we wanted to proceed with the objective of measuring the ID family of genes (ID1, ID2, ID3, ID4) and the NRF1 gene both groups known to be vital in redox oxidative stress repair in the brain in order to see if fold change values across AD cases versus controls would be significant. And if so, the next step would be to gather a list of all ID3 target genes that were found to be significant in DGE comparison analysis. Finally, the goal would be to discretize these significant genes and also the clinical variables of gender, disease, and age in order to find any causal network structures for novel pathways involved in AD pathogenesis. In the files attached, you can find the clinical information regarding the samples in this dataset such as the number of controls versus cases of AD patients on GEO at GSE104687. I plan to provide an update of ths analysis weekly to help any other individuals who seek to utilize this database or perform RNA-Seq analysis. Thank you and let me know of any questions you may need me to clarify.