Goal: to investigate the differences and similarities between sub-groups of patients with frontotemporal dementia (FTD), and their differences with the control group. Such differences were investigated across brain regions that are most affected by FTD.
Method: the data was processed in BASH using the in-house pipeline. The analysis was performed using the R and Bioconductor software tools.
- Model genome-wide count data (CAGEseq) as a function of case or control status
- Identified modules containing highly correlated genes using the weighted gene coexpression network analysis
- Functional annotation of the hits (differentially expressed genes or modules containing highly correlated genes) was performed using the hypergeometric test.
NOTE
-
The first three steps of the analysis generate the count table using the in-house pipeline. The relevant scripts are in the folder, "S1to3_Run_Pipeline_CAGEseq".
-
Then the annotation of the count table is performed using the scripts in the folder, "S4to5_Annotate_CounTable_CAGEseq".
-
The scripts relevant to the differential gene expression (DGE) analysis and the pre-DGE data preparation/crosscheck is here: " S6to7_preDE_DGEanaly_CAGEseq".
-
The functional annotation (of the significant DEG) using hypergeometric test was carried using the script at "S8_DEG_HypGeomtest_CAGEseq".
-
Finally, Weighted Gene Coexpression Network Analysis (WGCNA) was performed to find out coexpression modules and intramodular hub genes. Breifly, the stepwise scripts were organized into the following steps: S1-Data preparation and cleaning; S2-Network construction; S3-Correlation(ME, trait) and GS vs MM; S4: GS vs Intramodular-connectivity; S5-more network exploration; and, S6-Functional annotation of the interesting modules. The relevant scripts are at "WGCNA_StepwiseScripts_CAGEseq".
Used the following R packages:
- base
- CAGEr
- stringr; stringi; gdata
- openxlsx; xlsx; readxl
- GenomicFeatures; org.Hs.eg.db; BSgenome.Hsapiens.UCSC.hg38; TxDb.Hsapiens.UCSC.hg38.knownGene; AnnotationDbi; GO.db; Biobase; KEGG.db;
- ChIPseeker
- S4Vectors; Hmisc; reshape2; muStat
- edgeR; sva; statmod; HTSanalyzeR; snow; preprocessCore; GSEABase; WGCNA; impute; psych; gdata;
- RColorBrewer; pheatmap; plot3D; plotrix