Chapter 3 Overview of Omics Data Types
For each data type, include information on:
- Platform(s)
- Batches (and whether batch effect is significant)
- Any logistical (shipping, storage)/lab procedures conducted prior to collecting data (and indicate relevant technical parameters - ie. culture times, etc)
- Whether data was intentionally collected on specific subsets of donors
- Raw data processing pipeline
- Where you can find raw data files
- Missing value strategy, filtering, and normalization prior to statistical analysis
- Any additional notes
- Citations of papers that have published parts of the data
3.1 Gene expression (bulk RNA-seq)
RNA-seq profiles were collected starting in [YEAR]in batches of islets from ~10-20 donors. [Include information on batches]. Raw reads were aligned to the [BLANK GENOME] using [BLANK SOFTWARE version BLANK]. The counts matrix was filtered to remove any genes with fewer than [X] mean counts. The counts were normalized for sequencing depth and adjusted to account for high variance/high abundant genes using the [BLANK] method in the limma voom R package (version #). Batch effect was corrected using the Combat-Seq R package (version #). Since
3.2 Gene expression (Nanostring)
50 islets are hand picked in Edmonton. Cells are washed with PBS then collected in 100ul of RLT Beta-mercaptoethanol. Islets arestored at -80C before shipping to the Lynn Lab.
3.3 Protein expression (proteomics)
Human islets are shipped to Jim Johnson’s lab at UBC. 300 Islets are hand picked, washed with pbs and pellet is snap frozen and stored at -80C. Details of lysis and analysis are found here.