![]() Our system was able to classify seven morphological subtypes of Golgi-derived membrane structures, namely globules, lumps, loops, short, medium, long, and branch types with more than 96% accuracy. #CELLPROFILER MEASURE FIBER LENGTH MOVIE#Our system revealed a negative correlation between tubule and Golgi cisternae intensity prior to blink-out, supporting that tubular structures are derived from the Golgi apparatus, and that this can be quantified from live cell movie data. Our segmentation method was found to have the lowest root-mean square error (RMSE) compared to other segmentation methods. Correlation analysis is used to imply possible conversion between different Golgi tubule morphological subtypes. A bagged decision tree is chosen as a classifier for morphological subtypes. A total of 34 morphological features are used for classification by supervised machine learning. A combination of adaptive local normalization and Otsu’s thresholding methods are used for segmentation of subcellular objects. In this study, we demonstrate a semi-automated and user-friendly method, 2D-GolgiMorphSubtype system, to segment, classify, and quantify Golgi cisternae and Golgi-derived membrane structures. However, reliable quantification methods for Golgi-derived tubules in living cells have not been established. Quantification of such dynamics is crucial for further understanding the mechanism of regulation of this organelle. ![]() globules, networks, branches, and tubules. According to time-lapse imaging data, Golgi-derived membrane structures are highly dynamic and vary in number and shape, i.e. The exact physiological roles and the molecular mechanism of regulation of dynamics of such membranous structures are not fully understood. The Golgi apparatus plays a key role in the secretory pathway of eukaryotic cells. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |