Creating a Community by Integrating Expertise, Tools and Ideas
A personal reflection on the first NPLinker eScience workshop for Natural Product Discovery
This blog post is written by Wageningen University & Research Assistant Professor Justin van der Hooft who participates in the Netherlands eScience Center’s Open eScience Call 2021. His project, “A community-supported workflow connecting microbial genes, and organisms to their molecular products”, aims to facilitate the finding of novel bioactive molecules from nature. It does so by enabling the integrated use of various omics data types.
Motivation for the workshop
Nature is still full of unknown chemistry with possible beneficial properties such as antibiotic ones. Genomics technologies measure the genetic content of organisms in which the blueprint of their chemical output is present. Metabolomics technologies generate tons of spectral data of actually produced molecules. However, the analysis of large-scale omics datasets is still very challenging, especially when several types need to be integrated. What can we predict from genomics? What can we understand from metabolomics? And how best to connect these two workflows when doing exploratory data analysis without drowning in the data?
In order to find answers to these and other relevant questions together with early-career talented researchers, Marnix Medema and I organized the NPLinker eScience workshop.
Outline of the workshop
During this first workshop in a series of two, we aimed to bring together researchers interested in integrative omics analysis with various backgrounds in terms of their career stage, the technologies they are already familiar with, and the organisms they work on. Following a selection process, 30 participants were selected to come to Wageningen to learn about analyzing genomes and metabolomes, and to get an outlook towards the integration of the two. The aims of the workshop were:
- Bring together Genome & Metabolome Miners and learn the motivation and key characteristics of main genome and metabolome mining tools.
- Educate and train natural product researchers and other interested scientists in genome and metabolome mining.
- Establish together with the community what are the perceived main challenges in combining genomics and metabolomics, and the wishes for tools that address these.
To stimulate networking amongst the participants and instructors, several measures were put in place: extended coffee and tea breaks, the opportunity to present a poster, and group discussions in changing compositions. We are grateful for the financial support from the eScience Center to make the workshop affordable for participants from across the globe through a travel reimbursement scheme and to enable the presence of 4 international instructors.
Key insights
If there is one thing that became clear to me during the workshop, then it is this: multidisciplinary research requires open science and collaborative efforts to come to full fruition. During the workshop, one of the most heard perceived difficulties to do integrative omics analysis, is the (perceived) steep learning curve to get familiar with another type of omics data and the analysis workflow that comes with it. How to judge if this data is of good quality? How do you know if you used the right settings during the analysis? Hence, to do integrative omics analyses for natural product discovery, effective knowledge and skill transfer is needed. Here, having open access materials available is one part of the solution: researchers can use them to get familiar with the topic and the main things to consider for a successful project outcome, and teachers can use the materials to get inspired to teach this topic to students. We have also discussed the need for clear guidelines and instructions to run analysis software tools, and how the collaborative writing of documentation could be a solution to get to instructions that are clear(er) for a wide(r) audience. Finally, most tools, especially those relying on machine learning principles, perform best with increasing amounts of available data: all participants agreed that sharing is caring, and Cunliang Geng, research software engineer from the eScience Center, also provided an overview of how to make your data and software FAIR.
Practice as you preach
There was a lot of knowledge and skill exchange between participants themselves and participants and instructors that cannot be easily captured here. One tangible outcome for the workshop partipicants is the collaborative notes document that they collectively created and which is full of useful links and thoughts. Another tangible outcome of the workshop for the broader community is that the instructors all made as much as possible the materials available under an open access licence! I hope that this initiative will help to further foster teaching the future generation of omics miners, to aid in setting up teaching materials for genome mining, metabolomics, and metabolome mining, and to promote discussions in between and bring together the fields of genome mining and metabolome mining.
Personal reflection
Overall, I am glad that the participants generally liked the workshop format and content and that they look back at a great week in Wageningen. In particular, I was curious if the group discussions after lunch were appreciated, and based on the evaluation this was a good move, as it prevented the “after-lunch dip”, yet at the same time the participants could share their insights and frustrations. Furthermore, many of them were happy to realize that there are more scientists out there that struggle with integrating omics analyses and that have complementary skills and background to them. This was exactly the idea behind the community building! Also, we got some valuable feedback when organizing future workshops as this one: it is challenging to find the right balance for everyone between theoretical lectures and hands-on tutorials. Big thanks go out to all involved for their active participation, including the secretarial support in Wageningen. Finally, I look forward to organizing the second NPLinker eScience workshop that will be fully focused on how integrating genome and metabolome analysis can boost natural product discovery. The final day of the current workshop already demonstrated the currently ongoing developments within the NPLinker eScience project as well as by research groups around the world.
