paperKB
coga / coga-kb
Help
Sign in

Chunk #40 — Materials and Methods — Clustering of samples by brain region

Source
Abundant quantitative trait loci exist for DNA methylation and gene expression in human brain.
Embedded
yes

Text

Performing a Hierarchical Clustering (HCL) [9] of the sample profiles using the TM4 MeV version 4.1.01 tool [32], with Euclidian distances and ‘Average Linkage clustering’ resulted in the samples separating fairly well by brain tissue region. Where in mRNA separation of samples into 4 clusters matching brain tissue region was clear. The frontal and temporal cortices could not be separated in within CpG data. For clustering all detected data was used for mRNA; only detected data for autosomal probes was used in clustering of the CpG data, otherwise sub-clusters based on gender appeared. The HCL samples trees were saved a Newick tree files and plotted again using the HyperTree tool (http://hypertree.sourceforge.net/).