Self-supervised feature extraction from image time series in plant phenotyping using triplet networks.

MOTIVATION: Image-based profiling combines high-throughput screening with multiparametric feature analysis to capture the effect of perturbations on biological systems. This technology has attracted increasing interest in the field of plant phenotyping, promising to accelerate the discovery of novel herbicides. However, the extraction of meaningful features from unlabeled plant images remains a big challenge. RESULTS: We describe a novel data-driven approach to find feature representations from plant time-series images in a self-supervised manner by using time as a proxy for image similarity. In the spirit of transfer learning, we first apply an ImageNet-pretrained architecture as a base feature extractor. Then, we extend this architecture with a triplet network to refine and reduce the dimensionality of extracted features by ranking relative similarities between consecutive and non-consecutive time points. Without using any labels, we produce compact, organized representations of plant phenotypes and demonstrate their superior applicability to clustering, image retrieval and classification tasks. Besides time, our approach could be applied using other surrogate measures of phenotype similarity, thus providing a versatile method of general interest to the phenotypic profiling community. AVAILABILITY AND IMPLEMENTATION: Source code is provided in https://github.com/bayer-science-for-a-better-life/plant-triplet-net. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Selective Modification for Red-Shifted Excitability: A Small Change in Structure, a Huge Change in Photochemistry.

We developed three bathochromic, green-light activatable, photolabile protecting groups based on a nitrodibenzofuran (NDBF) core with D-π-A push-pull structures. Variation of donor substituents (D) at the favored ring position enabled us to observe their impact on the photolysis quantum yields. Comparing our new azetidinyl-NDBF (Az-NDBF) photolabile protecting group with our earlier published DMA-NDBF, we obtained insight into its excitation-specific photochemistry. While the "two-photon-only" cage DMA-NDBF was inert against one-photon excitation (1PE) in the visible spectral range, we were able to efficiently release glutamic acid from azetidinyl-NDBF with irradiation at 420 and 530 nm. Thus, a minimal change (a cyclization adding only one carbon atom) resulted in a drastically changed photochemical behavior, which enables photolysis in the green part of the spectrum.

A Study in Scaffold Hopping: Discovery and Optimization of Thiazolopyridines as Potent Herbicides That Inhibit Acyl-ACP Thioesterase.

In the search for new chemical entities that can control resistant weeds by addressing novel modes of action (MoAs), we were interested in further exploring a compound class that contained a 1,8-naphthyridine core. By leveraging scaffold hopping methodologies, we were able to discover the new thiazolopyridine compound class that act as potent herbicidal molecules. Further biochemical investigations allowed us to identify that the thiazolopyridines inhibit acyl-acyl carrier protein (ACP) thioesterase (FAT), with this being further confirmed via an X-ray cocrystal structure. Greenhouse trials revealed that the thiazolopyridines display excellent control of grass weed species in pre-emergence application coupled with dose response windows that enable partial selectivity in certain crops.