Exploring Omics Playground:
A Computational Biologist's Evaluation
Published on Nov 8, 2023
⏱ 8 min read
Our CEO, Murat Akhmedov, recently had the opportunity to interview Jonathan Landry, a computational biologist in the Genomics Core Facility at EMBL who supports the analysis of next-generation sequencing (NGS) data.
Jonathan Landry has an extensive background in cell biology, molecular biology, and sequencing, as well as experience in analyzing NGS data and applying computational methods to understand complex systems. Therefore, we wanted to get an expert opinion on Omics Playground and asked him to test drive the platform and share his feedback.
Here’s how the conversation went with Jonathan, where he talked about his practical experience testing the software and the potential impact of the platform on omics data analysis processes.
How would you describe Omics Playground to scientists who aren't familiar with it?
Jonathan: BigOmics provides a data analysis platform dedicated to omics data analysis and more precisely analyzing transcriptomics and proteomics datasets. The web-based platform allows you to start analyzing and visualizing your data in a code-free manner.
The first step is to upload your count tables and sample metadata to the system.
Second, the platform computes for you a set of QC metrics and pre-processes a set of defined analyses that include all standard approaches on such data types.
The user can then navigate into the different panels, to visualize, modify and start interpreting meaningful information extracted from the data.
Five classes of analysis are available, including looking at the clustering of your samples, their gene expression, in which pathways changes happen, how your groups of samples compare to each other or relate to external reference datasets such as drug signature responses or cell profiling of the sample of origin as an example. This analysis part is rendered in real-time and favors the interaction between the user and the platform.
What do you believe is the benefit of using Omics Playground for biologists, bioinformaticians, and sequencing units or CROs?
Jonathan: To my view, Omics Playground could address many challenges faced by different user types.
To start with, biologists with not enough coding expertise could be using the tools to grasp some meaningful information about their data. It is fast, user-friendly and very versatile. The implemented methodologies are standard in the field. It produces robust and transparent information about the processing and tools used. Plots are easily customizable and the user can interactively modify different parameters to export tables and plots in different sizes and formats.
Bioinformaticians could also offload some repetitive and standard workflows on the platform. Not all tools could be implemented and therefore the application might be restricted in its diversity. Nevertheless, new workflows and/or applications might be emerging on the Omics Playground in the future.
In a collaboration context, the recent implementation of dataset sharing allows quick access to the processed data and derivative plots with the team. It is very straightforward to communicate the main findings and enable the end-user to also start playing with the data.
What are the capabilities and strengths of the platform? What did you like the most?
Jonathan: I started my Omics Playground experience with bulk RNAseq datasets including 5 different protocols with 3 replicates each with 30 million counts per sample.
Between loading the entire dataset, computing and starting browsing the plots, it took about 20 minutes which is fast. It is fairly easy and intuitive to navigate through the different panels, customize the plots and export them in different formats. The different modules reassemble many essential analyses that one could undertake with this type of data.
One aspect that I found very interesting is the usage of different reference databases for the annotation of gene clusters or gene set enrichment of specific pathways. Integrating reference datasets or pathways, from established databases (Hallmark, MSigDB, KEGG, and Gene Ontology are some examples) is powerful and allows to corroborate the findings. Some specific features could also be found useful like the Drug Connectivity Map panel where gene expression signatures of known drugs could be interrogated and compared.
In addition to the extensive catalog of the analysis modules, it is important to notice the detailed documentation describing all the methods implemented in the backend of the dashboard. Some tutorial videos are also available as a quick introduction to how the platform is working for the rushing users.
What can be improved?
Jonathan: Although the documentation helps you to prepare the count tables, it might be challenging for biologists to get these counts in the first place. I understand that these mapping and feature counting steps are nowadays commonly delivered by sequencing facilities, but it might still remain a challenge for the users to prepare the input data.
To bridge this gap, it might be interesting to think about a module where one could upload the FASTQ files and necessary genome information (fasta and gene annotations) and get the data loaded into the platform after the mapping and quantification steps.
At BigOmics, we are committed to empowering researchers in omics data analysis. We continuously improve the user experience and ensure a smoother omics data analysis process. This year, we released a new version of Omics Playground, featuring a fresh-looking user interface, enhanced back-end, and new analysis modules. Some of the key improvements include:
- Improved interactivity and customization options for plots.
- Enhanced back-end computation speed.
- Introduction of new functionalities such as broader pathway analysis, gene subnetwork analysis (PCSF) module and the ability to share your dataset with your team or external collaborators.
We are actively working to expand the platform’s capabilities to make data preparation and uploading easier for new users and extend support for a wider range of species on the platform in the near future.
Furthermore, we are working on the interoperability of our platform with data storage and management software such as DNAnexus to provide our users with end-to-end analysis (link to press release) where they can start analysis from their FASTQ files.