Search for: All records

Creativity focuses on the generation of novel and useful ideas. In this paper, we propose an approach to automatically generating creative requirements candidates via the adversarial examples resulted from applying small changes (perturbations) to the original requirements descriptions. We present an architecture where the perturbator and the classifier positively influence each other. Meanwhile, we ensure that each adversarial example is uniquely traceable to an existing feature of the software, instrumenting explainability. Our experimental evaluation of six datasets shows that around 20% adversarial shift rate is achievable. In addition, a human subject study demonstrates our results are more clear, novel, and useful than the requirements candidates outputted from a state-of-the-art machine learning method. To connect the creative requirements closer with software development, we collaborate with a software development team and show how our results can support behavior-driven development for a web app built by the team. 

In this paper, we outline the need for a coordinated international effort toward the building of an open-access Global Ocean Oxygen Database and ATlas (GO 2 DAT) complying with the FAIR principles (Findable, Accessible, Interoperable, and Reusable). GO 2 DAT will combine data from the coastal and open ocean, as measured by the chemical Winkler titration method or by sensors (e.g., optodes, electrodes) from Eulerian and Lagrangian platforms (e.g., ships, moorings, profiling floats, gliders, ships of opportunities, marine mammals, cabled observatories). GO 2 DAT will further adopt a community-agreed, fully documented metadata format and a consistent quality control (QC) procedure and quality flagging (QF) system. GO 2 DAT will serve to support the development of advanced data analysis and biogeochemical models for improving our mapping, understanding and forecasting capabilities for ocean O 2 changes and deoxygenation trends. It will offer the opportunity to develop quality-controlled data synthesis products with unprecedented spatial (vertical and horizontal) and temporal (sub-seasonal to multi-decadal) resolution. These products will support model assessment, improvement and evaluation as well as the development of climate and ocean health indicators. They will further support the decision-making processes associated with the emerging blue economy, the conservation of marine resources and their associated ecosystem services and the development of management tools required by a diverse community of users (e.g., environmental agencies, aquaculture, and fishing sectors). A better knowledge base of the spatial and temporal variations of marine O 2 will improve our understanding of the ocean O 2 budget, and allow better quantification of the Earth’s carbon and heat budgets. With the ever-increasing need to protect and sustainably manage ocean services, GO 2 DAT will allow scientists to fully harness the increasing volumes of O 2 data already delivered by the expanding global ocean observing system and enable smooth incorporation of much higher quantities of data from autonomous platforms in the open ocean and coastal areas into comprehensive data products in the years to come. This paper aims at engaging the community (e.g., scientists, data managers, policy makers, service users) toward the development of GO 2 DAT within the framework of the UN Global Ocean Oxygen Decade (GOOD) program recently endorsed by IOC-UNESCO. A roadmap toward GO 2 DAT is proposed highlighting the efforts needed (e.g., in terms of human resources). 

Metagenomes encode an enormous diversity of proteins, reflecting a multiplicity of functions and activities. Exploration of this vast sequence space has been limited to a comparative analysis against reference microbial genomes and protein families derived from those genomes. Here, to examine the scale of yet untapped functional diversity beyond what is currently possible through the lens of reference genomes, we develop a computational approach to generate reference-free protein families from the sequence space in metagenomes. We analyze 26,931 metagenomes and identify 1.17 billion protein sequences longer than 35 amino acids with no similarity to any sequences from 102,491 reference genomes or the Pfam database. Using massively parallel graph-based clustering, we group these proteins into 106,198 novel sequence clusters with more than 100 members, doubling the number of protein families obtained from the reference genomes clustered using the same approach. We annotate these families on the basis of their taxonomic, habitat, geographical, and gene neighborhood distributions and, where sufficient sequence diversity is available, predict protein three-dimensional models, revealing novel structures. Overall, our results uncover an enormously diverse functional space, highlighting the importance of further exploring the microbial functional dark matter.