Bulk RNAseq Useful Resources (Living Document)

I am beginning my journey of simply doing bulk RNAseq (not single cell RNAseq) and I'd like to create a living document where I try to make your life easier by posting useful resources and step by steps on each part of the process

****Great high level resource for describing the various steps in RNAseq and why they are being done

Experiment

After performing the appropriate experimentation, generally lysis of the cells or tissue is done in RLT buffer (as provided by the Mini or Midi Qiagen Prep Kits). We generally supplement the RLT buffer with 1% Beta-mercaptoethanol. For cells, these are lysed directly in the cultured vessel (i.e. within each well of a 24, 12, 6 well plate). This is then left in lysis buffer for 10 minutes with the plate on ice and RNA transferred to microcentrifuge tubes for RNA extraction. RNA extraction is done via Mini/Midi Prep Kits followed by nanodrop to determine the concentration of the RNA.

Bioanalyzer (quality of RNA)

After lysis and extracting RNA, the next steps are to ascertain the quality of the RNA by BioAnalyzing! The Bioanalyzer (usually done on an Agilent Bioanalyzer platform) outputs an RNA Integrity Number (RIN) which ranges from 0-10. This number is essentially determined by an algorithm which informs you how intact and quality the RNA is. RIN >7-8 is generally considered to be “good” and sufficient for RNAseq.

*There are different chips available for this (nanochip or picochip), and thus you can get away with low concentration of RNA (as low as 5ng/ul) using a picochip.

Example Output with RIN 9.40

Sequencing parameters (For my use-case of simple differential gene expression)

Concentration for RNAseq: 12.5ng/ul in a total of 20uL for a total of 250ng
Type of Library: rRNA depletion (in my case). Illumina has a brief explanation of this (here)
Run Type: PE100 (PE = Pair end which sequences both ends of the fragment and gives higher quality data) (100 is the base pairs of the fragment)
Total Number of Reads per sample: 25M

All samples are then sent off to a core to get sequenced which takes approximately 4 weeks.

Raw Data

After 4 weeks the raw data is returned to you as a “fastq” file. I was also provided with a link to view some information on illumina's website. The first step is unzipping these files as they are often sent as a .gz file. For 12 samples, the storage size of the raw data was approximately 250 GB

%Q30 = The percentage of reads that came back with a quality score of 30 or higher.
- Detection of some reads can be an issue if there is dim fluorescence or if there is “low diversity” in a region on the flow cell and the machine is incapable of distinguishing between reads (see video above)

Data Processing

***USEFUL TOOL: GALAXY***

All the data processing can be done through this tool. I worked with a collaborator who processed the samples up until the step of running through “Kallisto” which generates abundances of transcripts. These files can be uploaded into Galaxy, after which I did the Differential Gene Expression using DEseq2 function on Galaxy. Then annotated the files afterwards on Galaxy.

Because I am using a human cell line, I annotated it using the “Homo_SapienGRCh38.109.gtf” as the reference genome

***USEFUL TOOL: http://geneontology.org/ ***

You can simply paste any relevant genes (i.e. those with most log2fc for example) and have GO enrichment analysis done to uncover which biological processes are associated with those set of genes.

To be continued (As of 7/7/2023)

Bulk RNAseq Useful Resources (Living Document)

Bulk RNAseq Useful Resources (Living Document)

Topics

DOI