Ten simple rules to cultivate transdisciplinary collaboration in data science

CyVerse: Cyberinfrastructure for open science

CyVerse, the largest publicly-funded open-source research cyberinfrastructure for life sciences, has played a crucial role in advancing data-driven research since the 2010s. As the technology landscape evolved with the emergence of cloud computing platforms, machine learning and artificial intelligence (AI) applications, CyVerse has enabled access by providing interfaces, Software as a Service (SaaS), and cloud-native Infrastructure as Code (IaC) to leverage new technologies. CyVerse services enable researchers to integrate institutional and private computational resources, custom software, perform analyses, and publish data in accordance with open science principles. Over the past 13 years, CyVerse has registered more than 124,000 verified accounts from 160 countries and was used for over 1,600 peer-reviewed publications. Since 2011, 45,000 students and researchers have been trained to use CyVerse. The platform has been replicated and deployed in three countries outside the US, with additional private deployments on commercial clouds for US government agencies and multinational corporations. In this manuscript, we present a strategic blueprint for creating and managing SaaS cyberinfrastructure and IaC as free and open-source software.

Inclusive collaboration across plant physiology and genomics: Now is the time!

Within the broad field of plant sciences, what are the most pressing challenges and opportunities to advance? Answers to this question usually include food and nutritional security, climate change mitigation, adaptation of plants to changing climates, preservation of biodiversity and ecosystem services, production of plant-based proteins and products, and growth of the bioeconomy. Genes and the processes their products carry out create differences in how plants grow, develop, and behave, and thus, the key solutions to these challenges lie squarely in the space where plant genomics and physiology intersect. Advancements in genomics, phenomics, and analysis tools have generated massive datasets, but these data are complex and have not always generated scientific insights at the anticipated pace. Further, new tools may need to be created or adapted, and field-relevant applications tested, to advance scientific discovery derived from such datasets. Meaningful, relevant conclusions and connections from genomics and plant physiological and biochemical data require both subject matter expertise and the collaborative skills needed to work together outside of specific disciplines. Bringing the best expertise to bear on complex problems in plant sciences requires enhanced, inclusive, and sustained collaboration across disciplines. However, despite significant efforts to enable and sustain collaborative research, a variety of challenges persist. Here, we present the outcomes and conclusions of two workshops convened to address the need for collaboration between scientists engaged in plant physiology, genetics, and genomics and to discuss the approaches that will create the necessary environments to support successful collaboration. We conclude with approaches to share and reward collaboration and the need to train inclusive scientists that will have the skills to thrive in interdisciplinary contexts.

Ten quick tips for sequence-based prediction of protein properties using machine learning

The ubiquitous availability of genome sequencing data explains the popularity of machine learning-based methods for the prediction of protein properties from their amino acid sequences. Over the years, while revising our own work, reading submitted manuscripts as well as published papers, we have noticed several recurring issues, which make some reported findings hard to understand and replicate. We suspect this may be due to biologists being unfamiliar with machine learning methodology, or conversely, machine learning experts may miss some of the knowledge needed to correctly apply their methods to proteins. Here, we aim to bridge this gap for developers of such methods. The most striking issues are linked to a lack of clarity: how were annotations of interest obtained; which benchmark metrics were used; how are positives and negatives defined. Others relate to a lack of rigor: If you sneak in structural information, your method is not sequence-based; if you compare your own model to "state-of-the-art," take the best methods; if you want to conclude that some method is better than another, obtain a significance estimate to support this claim. These, and other issues, we will cover in detail. These points may have seemed obvious to the authors during writing; however, they are not always clear-cut to the readers. We also expect many of these tips to hold for other machine learning-based applications in biology. Therefore, many computational biologists who develop methods in this particular subject will benefit from a concise overview of what to avoid and what to do instead.

Measuring the influence of non-scientific features on citations

Citations remain a prime, yet controversial, measure of academic performance. Ideally, how often a paper is cited should solely depend on the quality of the science reported therein. However, non-scientific factors, including structural elements (e.g., length of abstract, number of references) or attributes of authors (e.g., prestige and gender), may all influence citation outcomes. Knowing the predicted effect of these features on citations might make it possible to ‘game the system’ of citation counts when writing a paper. We conducted a meta-analysis to build a quantitative understanding of the effect of similar non-scientific features on the impact of scientific articles in terms of citations. We showed that article length, number of authors, author experience and their collaboration network, Impact Factors, availability as open access, online sharing, different referencing practice, and number of figures all exerted a positive influence on citations. These patterns were consistent across most disciplines. We also documented temporal trends towards a recent increase in the effect of journal Impact Factor and number of authors on citations. We suggest that our approach can be used as a benchmark to monitor the influence of these effects over time, minimising the influence of non-scientific features as a means to game the system of citation counts, and thus enhancing their usefulness as a measure of scientific quality.