The rapid growth of high-throughput omics technologies has created a need for integrated, user-friendly tools that enable robust and scalable analysis of complex biological data. In JMP Pro 20, we introduce the Genomics Starter, a structured and intuitive interface designed to streamline end-to-end workflows across genomics, transcriptomics, and microbiome studies.
Inspired by the familiar JMP Starter framework, the Genomics Starter organizes complex analytical tasks into a guided, menu-driven experience, allowing users to efficiently navigate key stages of analysis, including data import, preprocessing, and downstream modeling. The platform supports diverse data types (e.g., VCF, FASTQ, and wide-format matrices) and incorporates essential preprocessing steps, such as quality control, normalization, and encoding verification.
The Genomics Starter integrates advanced statistical capabilities, including distance matrix computation with multiple metrics and embedded PERMANOVA for group-based inference. It also supports exploratory and multivariate analyses through PCA, clustering, and interactive visualization, enabling users to uncover structure and variability in high-dimensional data sets. Flexible normalization approaches, including reference-based methods, accommodate a wide range of experimental designs, such as batch-structured metabolomics and microbiome data.
By combining statistical rigor with an accessible and guided interface, the Genomics Starter lowers the barrier to advanced omics analysis while maintaining analytical depth. This integration within JMP Pro provides a unified environment that enhances usability, reduces workflow fragmentation, and improves reproducibility, ultimately accelerating discovery in life sciences.
Hi, this is Farzaneh Rastegari, and I'm a Research Statistician Developer with the JMP Life Science R&D team, and my background is in computational genomics and bioinformatics. Today, I would like to show you one of the projects I have been working on recently, the Genomics Studio. Genomics Studio is a new upcoming add-in for JMP Pro that gives genomics researchers a single home base for omics analysis. If you have ever worked across genetics, RNA-seq, microbiome, proteomics, or single-cell data, you know the challenges. Every data type tends to pull you into a different tool, a different menu, and completely different workflow sometimes.
The idea behind Genomics Studio is that it doesn't have to work that way. Here, you open one add-in, tell it what data you're working with, and it takes you directly into a workflow designed for that data type with JMP's native statistical tools underneath, and Python or R available when needed. We organized Genomics Studio around six biological domains: genetics, RNA-seq, microbiome, single-cell, proteomics, and multi-omics. Each one has its own tile on the home screen, and it's the entry point into workflow design around that type of data. For example, if I go into genetics, I see tools for genotype data, quality control, imputation, population structure, marker association, and predictive modeling.
If I go into RNA-seq, a different workflow built around normalization, differential expression, and enrichment analysis. This is not just a collection of unrelated tools in one place. Each domain uses tools and terminology that makes sense for researchers working with that type of data. When you choose a domain, you are not locked into a rigid step one, step two, step three pipeline. What you see is a suggested workflow based on how that type of analysis is commonly done, but every step can also stand on its own. For example, if you have done your preprocessing on your data somewhere else and only want to run one analysis step, that's fine, too.
We designed it this way because real data analysis is not completely linear, and we didn't want to force users into a workflow they don't need. From anywhere in the app, you can type a keyword or tool name somewhere, something like PCA, or marker, or RNA-seq, or whatever by name and jump directly to that method by that. Right now, this is by keyword, sub-string matching, rather than a natural language question answering system. That's where it is today, and we have planned to make it smarter for the future.
The other thing is R is an optional connection. Most steps either run directly in JMP or call Python when needed, but connecting to R gives you access to add additional methods, especially through R or Bioconductor. But you don't need an R connection to use Genomics Studio, basically. In the screenshots, you can see again what it looks like. This is the first page. Here, you can see the application interface that you can see different domains here, and for example, search marker S here. You can see the marker statistics and marker simulation as an example of pulling up for the search. You can directly jump from here to those tools.
Here, you can see the workflow. For example, I'm in the genomics workflow. Here you can see the tool that each step is running by, for example, JMP for all of the steps here, the status of the step. For example, it's done for the import, and it's done for another population structure. You don't need to follow all the steps after each other. This one is in running status, actually. The expected input and output and the description for that tool is also here. Also, there is a link for the documentation.
For each domain, we have divided the workflow into import, preprocessing, and analysis, and the engine for running for that domain in this table. Every domain starts with the import, so there is a consistent entry point, but once you get into preprocessing and analysis, the workflows become more specific to the biology. For example, the genetics workflow here, after import, it has QC, quality control, impute, PCA, marker association, or predictive modeling. Or, for example, in the microbiome, after import, you have the quality control, CLR normalization, PCoA, UMAP, variance components, PERMANOVA, differential abundance, and response screening. The workflows are based on the domain here.
The overall architecture stays familiar, while the actual methods change depending on the type of data. One important point is that this table shows what's implemented today, but the coverage is still growing, and the architecture makes it possible to add any other methods afterward. Stepping back from the individual workflows, there are a few main ideas behind Genomics Studio. First, it gives you one workplace for multiple omics domains instead of having those workflows scattered across different places. Second, it is guided, but flexible, so you have a suggested path when you need guidance, but if you want to search a method or just do one step, you can do that.
Third, it works across JMP, Python, and R, routing each step to an appropriate engine if needed. Each workflow also links to the documentation and references, so you can see what the method is doing and where it comes from, actually. Finally, you can export an HTML report that someone else can open without needing the add-in or JMP installed, which makes it easier to share your work with collaborators. One small feature that actually I myself like it, it's the dark mode if you're interested. This is a new add-in under active development, and it's not in the JMP Marketplace yet, but it is coming soon.
If you work in any of these areas, and there is a method you need which isn't here yet, or if you have any questions for the workflow, any suggestions or questions about the workflows, we would like to hear your feedback. That actually helps us to go forward with that. If you have any questions or if you want to see any particular domain in detail and see a demo, find me at the JMP Discovery, and I would be happy to walk through it with you. That's it. Thank you for your attention, and I'm happy to answer any questions.
Presenters
Skill level
- Beginner
- Intermediate
- Advanced
