Genomics and wide fitting data in JMP Pro 17

[#1]

JMP Pro 17 incorporates a significant change to how genomic data analysis is performed in JMP. JMP previously relied on a backend connection to SAS to perform primary data import and computations using customized user interface dialogs and output dashboards in JMP. The new release allows you to do this work directly in JMP Pro and no longer uses SAS or additional customizations. Distinguished Research Fellow Russ Wolfinger (@russ_wolfinger) says, “This dramatic change will allow streamlining and refining of many genomics analysis workflows. While much work remains, the initial offering is very powerful and offers many key capabilities.” Russ gives us a look at these exciting changes to genomics data analysis.

What problems do the changes solve for our users?
The new framework relies directly on the standard JMP interface, making many steps much more familiar and straightforward for those who know JMP. Additionally, a new guide, Genomic Analysis with JMP Pro, will assist users who are working with these types of data and analyses. 

Can you give an example of how the experience might be for a JMP user?
The typical starting points are File > Open for data in a single file or File > Import Multiple Files for data from several files. Many common formats such as vcf, bed, gff, and IDAT are automatically detected. The import routines typically create a JMP table in “wide” form, with samples as rows and genes as columns, and this wide table is ready to go for use in many JMP Pro platforms, depending upon analysis objectives.

Many of the data table operations and analysis platforms have been optimized for these “wide data” problems, allowing the user to easily apply JMP Pro’s analysis tools with many thousands of variables. The release also features three new platforms that are particularly relevant: Marker Statistics, Marker Simulation and Multivariate Embedding. 

What are the impacts this can make on users’ work and on organizations?
With key new work to improve performance on large wide tables, several import routines and analyses are orders of magnitude faster. Results from each step are much more immediate, permitting more ready interpretation and possibilities for backtracking, side-by-side comparisons, and rapid convergence to important discoveries and conclusions.

What do you think will surprise or delight users about this?
Many users will be surprised at how fast and powerful the analysis of detailed genomic data analysis is within the new JMP Pro framework. We have also created a companion online book, Genomic Analysis with JMP Pro, to guide users through applying JMP Pro to these types of problems.