0. HamidLab routines

Accessing Clarke and RStudio server

All staff and students can access Clarke (our workstation) via terminal using ssh:

ssh user@workstation

or

ssh user@10.0.1.25

where user can be either staffs or student.

If there is a need to use graphical user interface (GUI), use ssh with a -Y flag instead:

ssh -Y user@workstation

Upon login from a new device, you need to set up SSH authentication keys. Do refer to this tutorial on how to do so.

You also have access to RStudio server via the address 10.0.1.25:8787. Input your user and password when prompted.

Add remote server location to Ubuntu File Manager using SSH

Open Ubuntu's file manager. At the bottom of the left side panel, click on + Other Locations. There is a Connect to server input field at the bottom. Enter the following changing user accordingly.

ssh://user@10.0.1.25

Now press Connect. Within a few seconds, Ubuntu's file manager will mount the server location and will be visible on the left side panel.

CDN-BC projects

All CDN-BC projects reside on the RAID drive at the following path: /media/cdn-bc/RAID/Projects2.

Project ID and name

Each folder in the Projects directory will be named according to the following nomenclature:

ProjectID_CollabName_ShortDescription

ProjectID: As of April 2023, project IDs will be prefixed with the initial 'BC' followed by a 3-digit numeric (e.g. BC001, BC002). Imaging projects will be prefixed with the initial 'BCI' instead. The first project ID will begin with the number '001' and this number will increment by 1 for subsequent projects. ProjectID will serve as our internal reference and will be used to reference reports and emails.

CollabName: First name of main collaborator. Keep it short.

ShortDescription: A 1-word description of the project (e.g. HgMmscRNAseq, DRerioIR)

Project directory structure

All project directories will adopt the following file structure:

├─ metadata
├─ originals
├─ outputs
└─ readmes

Each sub-directory serves specific purposes:

'metadata': contains miscellaneous project files including samples' metadata.
'originals': contains RAW data files such as FASTQ files or downloaded expression data.
'outputs': contains processed data stored neatly in folders.
'readmes': contains reports of the project's analysis in the form of Quarto/Markdown (preferred) or scripts.

Other folders can be made in the project directory as needed.

Command to create Project directory

The above structure can be created automatically in the Projects directory using the following command:

BCinit ProjectID_CollabName_ShortDescription

Centralised genomes

Frequently used reference genomes and transcriptomes are stored at /media/cdn-bc/RAID/Genomes/. This is the structure of the Genomes folder:

Genomes
└─ Organism or built
└─ Source or version
└─ FASTA, GTF or indices

Please keep this folder organised and updated!

Reports

The results of the analysis will be reported in the form of Quarto HTML documents. More information on Quarto and how to format Quarto documents can be found in this link.

A report template for the CDN-BC has been created and can be accessed on RStudio Server (TO BE TESTED) by clicking on:

File > New File > R Markdown...

Choose "Report (Qmd)" under the "From Template" section.