Computational analysis of single-cell/cytometry data
As the number of parameters that we can measure in cytometry increases, so too does the complexity of the data we are able to generate. As such, manual methods of analysis (such as 'gating') are not always suitable for the analysis of these datasets. Various computational tools can be harnessed to aid in the analysis of these datasets, but the use of these tools needs to be well considered.
Resources and software
Sydney Cytometry provides our users with access to physical analysis computers. Additionally, we provide our users with access to Virtual Research Desktops (VRDs), which can be accessed remotely. A VRD enables users to utilise significant computing power for analysis that might not be possible on local computers.
Sydney Cytometry also facilitates access to an number of graphic user interface (GUI) software options to facilitation computational analysis of cytometry data, such as FlowJo and cytofkit (open source, in R). Additionally, we provide access to coding software and scripts for computational analysis, such as R and R Studio, Python, and Matlab.
ANALYSIS Protocols and workflows
We have a variety of protocols and workflows available to facilitate data management and preparation, clustering (e.g. FlowSOM, PhenoGraph), dimensionality reduction (e.g. tSNE, UMAP), plotting, and visualisation. Many of these protocols we execute using R, including in our analysis pipeline ‘CAPX’, but we also provide protocols for many of these in other programs, such as FlowJo.
Many of our computational analysis workflows are run in R, and a number of helpful R scripts have been provided at www.github.com/sydneycytometry. Instructions and tutorials of many of these scripts are provided on the same page. We also have an introductory tutorial to using R in RStudio here: using R scripts in R studio.
CAPX (Cytometry Analysis Pipeline for the analysis of large and complex data): a script/workflow that facilitates the analysis of very large (<30 million cell) flow or CyTOF datasets through data management, clustering (using FlowSOM), and dimensionality reduction (using tSNE/UMAP). More information can be found here.
tSNEplots: a script to automatically generate tSNE plots from your data, with cells coloured by the expression of each marker in your panel.
CytoTools: a collection of scripts that are helpful for managing and analysing cytometry data, including generate summary data, annotating clusters in CSV/FCS files, and generating heatmaps.
AutoGraph: a script to automatically generate bar graphs/dot graphs with statistical comparisons overlaid.
REGULAR TRAINING SESSIONS
We provide regular training sessions in basic and advanced computational data analysis, which registered users can request at the link below.