By now, many of you have seen the recent announcement that our current JMP Genomics product will no longer be sold to new customers after March 31st. This decision has led to several questions internally, and as customers hear this news, we expect that they will have similar questions. Here are some of those questions along with answers from development and product management.
Question: Are we abandoning our JMP Genomics customers? Are "-omic" data analysis methods and tools still an important part of our product vision and roadmap?
Answer: No, we are not abandoning our customers, and yes, we fully intend to support our Genomics and related life sciences customers better than ever – but now with JMP Pro and not a separate product dedicated to genomics. The field of genomics is growing in use and importance, beyond just academic or basic research, and we expect that our current and future customers will want to use our products to solve important problems in this area.
Question: Why are we making this move right now, and so fast?
Answer: Improvements planned for JMP Pro 17 will provide an advanced level of capability and performance that allows it to stand on its own without the need for SAS to provide additional analysis capabilities. In addition, we must now forge our own path and we will no longer provide SAS products as part of any of our JMP product offerings after this year. At the same time, plans were already in place to implement a transition to make our Life Sciences products solely use JMP. With all these considerations, this is the best time to make this decision.
Question: What will be some of the benefits of this transition?
Answer: The previous JMP Genomics product had a JMP front end and used SAS as a back end, and because of that had some drawbacks:
- It was hard to install because it required a large installation package (12 GB), primarily because of the SAS components. Now, with Genomics inside JMP Pro, it is easy to download (less than 1 GB) and fast to install.
- Genomics users had to order a special version of JMP. Now they will have the features they need inside JMP Pro – there will be no need to obtain a different product or add-on.
- SAS was required in previous versions because JMP did not have all the features needed to handle Genomics data, including import and some statistical methods. Now JMP can do almost all these things without needing SAS.
- Using SAS data sets on disk suffered from slow disk access, large uncompressed files, and a mismatch between SAS and JMP files. Under JMP Pro the files will be much smaller, in compressed form, directly accessible to all platforms, and all the analysis will be in-memory, much, much faster.
- Previously, the files often had to be stacked into "tall" form, with few columns and a huge number of rows. The tall form was wasteful, with subject-data repeated across all the elements from that subject. It is much better in "wide" form; not only is it much more compact but utilizes many new features in JMP Pro to make handling wide data fast and easy.
- We have removed the bottlenecks in JMP Pro 17 that previously made handling hundreds of thousands of columns slow. The commonly used routines are now multithreaded and optimized for wide problems, sometimes hundreds of times faster than before.
Question: What capabilities and features will there be in JMP Pro 17 that meet the needs of current Genomics users and that will attract genomic data scientists?
Answer: JMP Pro 17 will be one of the only advanced analytics software packages providing the combination of an interactive and engaging user experience that allows for rapid point-and-click exploration of genomics data, advanced multivariate and predictive modeling tools, and a flexible and adaptive platform (through JMP Scripting and integration with other data science tools). New capabilities in JMP Pro for this type of work include:
- Performance and output enhancements to several platforms (including Response Screening, Principal Components, Fit Model, Fit Mixed, Hierarchical Clustering, K-Means Clustering) to better handle large, wide tables
- A new Genetics submenu with two new platforms: Marker Statistics and Marker Simulation
- Multivariate dimension reduction and summarization methods, including t-SNE and UMAP, in a new Multivariate Embedding platform
- Streamlined import of Illumina IDAT files
- New detailed training materials providing numerous tips, strategies, and common genomics analysis workflows
Traditional tools for analyzing genomics data tended to be programming-based, requiring a long learning curve to learn the programming language. With JMP Pro everything is interactive, point-and-click, walk-up-and-use easy. All the graphs are completely interactive, and analysis becomes a voyage of discovery, following clues and pursuing insights from the graphs. Contrast this with the plan-ahead learning-burdened script-based work process that dominates the process with open-source tools. With JMP, researchers can concentrate on the subject matter, rather than spending all their time fighting the tools of analysis. For advanced users that already have those programming and script-based skills and tools, JMP Pro will also be beneficial to them, giving them the ability to do rapid interactive analysis so that they can quickly understand their data and enabling them to better share their insights with others.
Question: Will JMP Pro 17 analysis of genomics data have the same user experience, or will it be different?
Answer: The style of operation of JMP Pro 17 for Genomics users will be different than the JMP Genomics product: instead of reshaping and characterizing your data so that it fits a pre-determined analysis workflow, users will be able to directly access the easy-to-use tools in JMP Pro and craft their own custom workflows.
Any additional questions?
If you have any other questions, feel free to comment below or send them to Sam Gardner (@SamGardner) in JMP Product Management.
@wendytseng I am not sure of the exact items that were optimized for many columns. Maybe @SamGardner would be able to find out from development.
Regarding the following benefit...is this only in Pro? For example, would one expect different performance in Response Screening in JMP vs JMP Pro all else equal?
"We have removed the bottlenecks in JMP Pro 17 that previously made handling hundreds of thousands of columns slow. The commonly used routines are now multithreaded and optimized for wide problems, sometimes hundreds of times faster than before."
@wendytseng , there is not a list that I know of. It is better to point out what we can do and if what a user wants is not there but was something they did do in JMP Genomics, then put it on the Wish List. There are a lot of capabilities that are not in JMP/JMP Pro but were in JMP Genomics. The way in which JMP Genomics approached the data can be done in JMP/JMP Pro, but take multiple steps with a few different capabilities combined.
Which capabilities that were in JMP Genomics that are not in JMP Pro? (Let me know if this info exists elsewhere.)
We will not have a separate Genomics menu or Genomics Starter for JMP Pro. There are two new platforms, Marker Statistics and Marker Simulation, which be in a separate submenu menu under the Analyze.menu.
Is there Genomics menu on new JMP Pro, instead of Genomics Starter of current JMP Genomics ?
@DonMcCormack JMP Pro will come with all of these platforms and features.
Will the new JMPG platforms be available to all JMP Pro users?
