A qualitative assessment of using ChatGPT as large language model for scientific workflow development.

BACKGROUND: Scientific workflow systems are increasingly popular for expressing and executing complex data analysis pipelines over large datasets, as they offer reproducibility, dependability, and scalability of analyses by automatic parallelization on large compute clusters. However, implementing workflows is difficult due to the involvement of many black-box tools and the deep infrastructure stack necessary for their execution. Simultaneously, user-supporting tools are rare, and the number of available examples is much lower than in classical programming languages. RESULTS: To address these challenges, we investigate the efficiency of large language models (LLMs), specifically ChatGPT, to support users when dealing with scientific workflows. We performed 3 user studies in 2 scientific domains to evaluate ChatGPT for comprehending, adapting, and extending workflows. Our results indicate that LLMs efficiently interpret workflows but achieve lower performance for exchanging components or purposeful workflow extensions. We characterize their limitations in these challenging scenarios and suggest future research directions. CONCLUSIONS: Our results show a high accuracy for comprehending and explaining scientific workflows while achieving a reduced performance for modifying and extending workflow descriptions. These findings clearly illustrate the need for further research in this area.

Adaptive camouflage: what can be learned from the wetting behaviour of the tropical flat bugs Dysodius lunatus and Dysodiusmagnus.

The neotropical flat bug species Dysodius lunatus and Dysodius magnus show a fascinating camouflage principle, as their appearance renders the animal hardly visible on the bark of trees. However, when getting wet due to rain, bark changes its colour and gets darker. In order to keep the camouflage effect, it seems that some Dysodius species benefit from their ability to hold a water film on their cuticle and therefore change their optical properties when also wetted by water. This camouflage behaviour requires the insect to have a hydrophilic surface and passive surface structures which facilitate the liquid spreading. Here we show morphological and chemical characterisations of the surface, especially the cuticular waxes of D. magnus Scanning electron microscopy revealed that the animal is covered with pillar-like microstructures which, in combination with a surprising chemical hydrophilicity of the cuticle waxes, render the bug almost superhydrophilic: water spreads immediately across the surface. We could theoretically model this behaviour assuming the effect of hemi-wicking (a state in which a droplet sits on a rough surface, partwise imbibing the structure around).  Additionally the principle was abstracted and a laser-patterned polymer surface, mimicking the structure and contact angle of Dysodius wax, shows exactly the behaviour of the natural role model - immediate spreading of water and the formation of a thin continuous water film changing optical properties of the surface.