Anatomics MLT, an AI tool for large scale quantification of ultrastructural traits

Abstract

The ever increasing breadth of biological knowledge has led to recent efforts to combine information from various fields into cell- or tissue atlases. Anatomical features are the structural basis for such efforts, but unfortunately large scale analysis of subcellular anatomical traits is currently a missing feature. Similarly, small phenotypic alterations of organelle- or cell-specific anatomical traits, such as an increase of the total volume or the number of mitochondria in response to certain stimuli, are currently hard to quantify. To provide tools to extract quantitative information from available 3D microscopic datasets generated with methods such as serial block face scanning electron microscopy we a) developed much improved fixation and embedding protocols for plants to drastically reduce processing artifacts and b) generated an easy-to-use AI tool for quantitative analysis and visualization of large-scale data sets. We make this tool available as open source.

Plant phenotyping relevance

植物の3D顕微鏡データから細胞・細胞小器官の構造形質を大規模定量するAIツールを開発しており、植物向け試料調製法も改良しているため、表現型取得・解析手法が研究の中心である。

Code and data availability

公開論文であることは確認できましたが、現在の公式API・許可済み取得経路では本文を自動取得できませんでした。

No evidence-backed public reproduction asset is currently recorded.