English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent advances in computer vision and robotics enabled automated large-scalebiological image analysis. Various machine learning approaches have beensuccessfully applied to phenotypic profiling. However, it remains unclear howthey compare in terms of biological feature extraction. In this study, wepropose a simple CNN architecture and implement 4 different representationlearning approaches. We train 16 deep learning setups on the 770k cancer cellimages dataset under identical conditions, using different augmenting andcropping strategies. We compare the learned representations by evaluatingmultiple metrics for each of three downstream tasks: i) distance-basedsimilarity analysis of known drugs, ii) classification of drugs versuscontrols, iii) clustering within cell lines. We also compare training times andmemory usage. Among all tested setups, multi-crops and random augmentationsgenerally improved performance across tasks, as expected. Strikingly,self-supervised (implicit contrastive learning) models showed competitiveperformance being up to 11 times faster to train. Self-supervised regularizedlearning required the most of memory and computation to deliver arguably themost informative features. We observe that no single combination of augmentingand cropping strategies consistently results in top performance across tasksand recommend prospective research directions.

Comparing representations of biological data learned with different AI paradigms, augmenting and cropping strategies