REFERENCE:
https://doi.org/10.1101/2020.12.08.415562
https://1drv.ms/b/s!Ath3f5ykGji3mi8xidQEgV1zLX-W?e=FSwVfm
Abstract
- benchmark 基準
- cell-to-cell variation
- locations, amounts, and variation of the 25 cellular structures.
- stereotypy: a structure’s individual location varied
- concordance: the structure localized relative to all the other cellular structures
Introduction
- sub cellular components and processes in space
- four orders of magnitude.
- global cell behaviors:
- protrusion 突出物 of a cell front
- retraction of a rear during cell migration
- daunting 令人退縮的 immensely complex 及其複雜的
- enormously 巨大的
- quest 尋求
- This baseline should represent the typical, or mean, cell within the population(種群), as well as the full range of normal variation of the population itself.
- unprecedented 前所未有的
- a fundamental benchmark for comparision with future analyses of cell shape and cell organization for cells in different states.
Result
- quality control workflows to create the Allen Cell Collection
- organelles 細胞器(包括綫粒體葉綠體什麽的)
- cellular structures
- compartments 間隔
-
consistently 始終如一的
- nucleoli DFC 致密纤维组分, the dense fibrillar component
- nucleoli GC the granular component
-
nuclear speckles
- cohesin 粘连蛋白
- histones
- nuclear envelope 细胞膜
- nuclear pores
- …
Methodology P33
registration
- centroids 形心
- preserving
- the biologically relevant, apical-basal（以顶点为基准的） axis (顶点的那个z轴危中心)
- the longest axis of the cell perpendicular(垂直的 ) to that axis
SHE -> SHAPE SPACE MAP P11 (RED PLACE）
- A method to “morph” all of the locations of all of the points within a cell into the idealized reconstructed cell shape the best represents the cell shape.
QUANTIFICATION AND STATISTICAL ANALYSIS P48
Building the cell and nuclear shape space P54
identifying the primary modes of shape variation
I didn’t understand the PCA usages, I can not get the real PCA for every cell. 24-2-2021 The following fucking mapping methods are derived by my self, but I cannot get the real inverse, the result SHcPCA are so weired.
- SHE + PCA
- you know what is PCA right?
- PCA +- delta depend on the shape space map distribution
- Probability density distribution of principal component values
- see picture1 in experiment section
Explanation
- MAXIMUM POINT IN PCA ( SHAPE SPACE) TO DO CLUSTERING
Experiment
- correlation between original image and reconstruction image
- reconstruction use nearest neighbor interpolation
- step by step, voxel by voxel ( not regular filling)
- correlation
- voxelwise pearson correlation between original image and recontruction image
- choose max point, shape model and shape space build:
- shape model 6
- As an example, the bin corresponding to the map point (0,0,0,0,0,-1.5σ,0,0) is highlighted in blue.
see suplementary materials(picture resources):
https://www.biorxiv.org/content/10.1101/2020.12.08.415562v1.supplementary-material
ANALYSIS
PCA shape
- in this dataset, the total height of the cell is largely independent of its overall volume. (supported or to support ‘Statistical Analysis of ‘)
