Getting Started

Obtaining a Katana account

To apply for a Katana account you can send an email to the UNSW IT Service Centre giving your zID, your role within UNSW and the name of your supervisor.You will be assigned storage space in two locations:

/home/zID (10GB limit)

/srv/scratch/zID (128 GB limit).

If your lab/group has a dedicated scratch space, they can contact us to grant you access to this shared space. This can be useful if your data and results files exceed you personal account space.

Using Katana servers.

Katana servers can be accessed via ssh at the following addresses:

katana.unsw.edu.au for Katana nodes used to start jobs and monitor jobs

kdm.restech.unsw.edu.au for the Katana Data Mover (i.e copying/moving files)

You can log into either using your zID and password.

Katana on demand is a simple way of accessing the Katana servers in a web browser, which can be accessed here.

For moving large data files to/from Katana, such as sequencing files, it is recommended to use a client software such as Filezilla. 

For more details on using  Katana see the official documentation 

Sequencing data can be specified as either:

• A path to a directory on Katana (eg. /srv/scratch/zID/myProject/Data).

• A link provided by the Ramaciotti Centre 

If your sequencing data has been generated at the Ramaciotti Centre, you will receive a url for downloading the data, such as: 

https://mydata.ramaciotti.unsw.edu.au/s/PiE96tn58FZAZbs/download/ProjectID.tar. You can provide this link in the nf-core pipeline, without downloading the data. 

• An NCBI SRA number 

If your would like to analyze publicly available data from NCBI/SRA, you can provide the SRA project ID, such as SRPXXXXXX or PRJNA XXXXXX without downloading the data. Note that GEO Id (GSEXXX) numbers are not supported. 

To run a pipeline you will need storage space for:

1. Input fastq files. We recommend you place your fastq files in your personal or group scratch space directory, and move them to the UNSW data archive after running the pipeline.

2. Output files. NextFlow pipelines generate large intermediate files, which allows the pipeline to resume without losing its progress. For example, the RNA-seq pipeline output is typically ~ 4-6 times larger than the input data (eg. if your input fastq files are 10Gb you can expect the output folder to take 40-60Gb. Most of this output consists of intermediate files, generated in the ./work sub-directory, which can be deleted after the run is complete.

Since most individual accounts don’t have enough storage space to run NextFlow pipelines on large sequencing data, we provide temporary storage in a dedicated scratch directory: /srv/scratch/seqflow.

You should create a sub-directory named as your zID:  /srv/scratch/seqflow/zID for your own use.This folder will be only accessible with your zID and password.

Important note: Data are automatically deleted from /srv/scratch/seqflow after 3 days of not being accessed, so you should copy your results files to you personal or group scratch directory after the pipeline successfully completes.

nf-core pipelines include extensive documentation (eg. nf-core RNA-seq). In addition, we strongly encourage you to spend some time reading the documentation of each software tool included in the pipeline.This includes the parameters of each tool, and the expected output files.