3.6.3 Specific Aims

Back to Phase II Grant Documents

Flow cytometry is used to rapidly gather large quantities of data on cell type and function. The manual process of classifying hundreds of thousands of cells forms a bottleneck in diagnostics, high throughput screening, clinical trials, and large scale research experiments. The process currently requires a trained technician to identify populations on a digital graph of the data by manually drawing regions. As the complexity of the data increases, this gating task becomes more lengthy and laborious, it is increasingly clear that elimination of human processing is essential to increasing both throughput and consistency. In clinical tests, and diagnostic environments, automated gating would eliminate a complex set of human instructions and decisions in the Standard Operating Procedure (SOP), reducing error and speeding results to the doctor. In many cases, the software will be able to order additional tests before the doctor has an opportunity to look at the first report. Currently no software is available to perform complex multi-parameter analyses in an automated and rigorously validated manner. FlowDx will fill an important gap in the evolution of the technology and pave the way for ever larger phenotypic studies, and translation of this research process to a clinical environment.

Technological innovation: Each customer's situation will dictate the level of security, the clustering algorithm(s), data integrity checks, and their own quality control tests for each assay. Customization is accomplished by means of the template technology available in the underlying FlowJo architecture. Currently in FlowJo, an established template takes new data and analyzes it first by grouping samples according to common characteristics, and then by applying fixed gates previously created on a typical data set. The template can also apply a range of statistical analyses to the resulting populations and arrange the results into tables and graphical layouts for examination by investigators.

FlowDx will build on this foundation by:
Validating the content and structure of incoming data
Enabling new gates created algorithmically and optimized in response to each new data set
Testing results against established values to verify performance and to label results with diagnostically useful information
Creating and recording reports for examination by investigators
Protecting the privacy of patient information throughout the process by using PGP's [4.] suite of authorization/encryption services in compliance with FDA 21 CFR Part 11 guidance

1. Develop the experimental protocol whereby a researcher can compare two or more classifications of identical datasets to study the differences, biases, and effectiveness of human and algorithmic classifiers.
Milestone: An abstracted workflow document beginning with a client's assay and finishing with automated analysis.

2. Describe and evaluate metrics that compare the performance of classification algorithms.
Milestone: Database utilities that apply four comparison metrics to the popmask and consensus files in the FlowDx database and records the results. Test documentation and training materials corresponding to this process.

3. Conduct analytical experiments on our identified use cases illustrating the potential of this technique to affect clinical analysis.
Milestone: Completed matrix of results showing which algorithms perform best for each use case and each comparison metric.

4. Iteratively implement the tools to automate these experiments, improve the experimental capabilities, and collaborate in new use cases.
Milestone: Operational implementations of the database, repository, and utilities.

Commercial Application: Many tools exist for manipulation of data with encoded formulae, but these require sophistication in the deployment of statistical computer languages and code "packages" that represent an unacceptably steep learning curve for many working labs. To date, there are few examples of automation in commercial flow cytometry analysis software. The current commercial version of Tree Star's software, FlowJo, can perform clustering automatically as part of a template analysis operation[2.], but this clustering is limited in scope and not independent of operator interaction.

Tree Star's collaborating scientists are eager to have automated analysis available for their use cases, and they represent a tiny fraction of the demand.

"Flow cytometers are widely found in all leading biomedical research institutions and universities where they are used for performing tasks that require analytical precision and high throughput. In addition, flow cytometers have a key role in hospital and medical centers worldwide, where they are widely used for diagnosis as well as research. There are several thousand flow cytofluorometers in clinical use worldwide." [3.]