Automatically Identify Predictive Biomarkers with Our Newest Algorithm
Why do some cancer patients respond well to immunotherapy but others do not?
Why do some people have a slow and painful recovery from surgery but others have a speedy recovery?
What are the biomarkers that can help predict these outcomes ahead of time?
High-dimensional single-cell analysis approaches are excellent for investigating such questions because many mechanisms of disease may only be visible at the single-cell level, eluding bulk analysis techniques. Emerging technologies such as high parameter fluorescence and mass cytometry, and powerful data analysis platforms like Cytobank, are providing unprecedented resolution for measuring single-cell biology. With these new technologies, a variety of specific cellular populations can be simultaneously identified, and anomalies can define different clinically relevant cohorts and serve as predictive diagnostics and prognostics.
The challenge of going from high-dimensional data to these useful findings lies in the analysis, which is often cumbersome, manual, subjective, and irreproducible. Our new version of the CITRUS algorithm aims to change that.
CITRUS (cluster identification, characterization, and regression) is now available in Cytobank Enterprise and Cytobank Premium. The CITRUS algorithm enables automated discovery of statistically relevant stratifying biomarkers within single-cell datasets containing numerous samples across multiple known endpoints (e.g., responders versus non-responders). In collaboration with Stanford’s Nolan Lab, we’ve integrated the algorithm to work seamlessly with Cytobank.
CITRUS can be used in any analysis requiring the comparison of groups of samples. Examples could include clinical comparisons, biomarker discovery, and any application relevant to building predictive or correlative models for the classification of samples in other independent data sets.
How it Works:
CITRUS automates multiple steps, using regularized supervised learning algorithms to best predict which endpoint group any given sample belongs to, or uses correlative methods for a correlative measure of the same. The results of a CITRUS run are clusters and associated features that differentiate the observed endpoint of the samples.
The Cytobank CITRUS Advantage:
- Get Started in Minutes
No need to know R, and no need for additional Computational Biology resources. The CITRUS tool is integrated seamlessly with Cytobank, and is easy to use.
“CITRUS is very easy to use and integrated into the Cytobank platform, which makes handling and organization of raw data possible. All of the algorithms (SPADE, viSNE, and CITRUS) are implemented in a way that allows easy execution of workflows.”
Institute of Medical Virology
University of Zurich
- Improve Efficiency and Accuracy of Your Data Analysis.
Instead of a raw data dump that is then subject to interpretation, CITRUS delivers automated identification of statistically-significant biomarkers.
- Gain the Cloud Advantage in Productivity and Data Integrity
Leverage the speed and efficiency of our high-powered servers in the Cytobank Cloud to gain results in a fraction of the time. Unlike slower, local plugins that require dedicated memory and limit your ability to multi-task, Cytobank’s CITRUS analysis can run imperceptibly in the background while while you use other parts of Cytobank, or exit the platform entirely. By storing in the cloud, your advanced analysis results remain integrated, well-managed for quick recall, and secure collaboration and preservation throughout time.
“With our CyTOF studies and analysis we often have the impression of assembling a railroad track right in front of the train, and incorporating CITRUS has helped us build the track out faster and further ahead than before. We can now do in two weeks what would otherwise take months of slicing populations into smaller groups to conduct the analysis”
— Patrick Reeves Ph.D.
Team Leader, Vaccine & Immunotherapy Center
Massachusetts General Hospital
Learn More About CITRUS:
Visit Our Support Site for Full Documentation
Sign Up for our CITRUS Webinar on Wednesday, October 19: