From f9b0d521e53ab425fd15b62d4fbee4ee2b689dcf Mon Sep 17 00:00:00 2001 From: Christian C Date: Wed, 13 Nov 2024 21:28:44 -0800 Subject: Structure README --- README.md | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) diff --git a/README.md b/README.md index 03ea18d..ec6cb76 100644 --- a/README.md +++ b/README.md @@ -1,3 +1,41 @@ # Representation of High-Dimensional Cancer Cell Morphodynamics in 2-D Latent Space [Paper - TBD](https://void) + +TODO: Fill in figures + +## Overview + +### Figure 1 + +![Latent Representation Pipeline](figures/Figure1.png "Latent Representation Pipeline") + +### Examples + +#### Spheroid + +![Spheroid Invasions](figures/Figure1.png "Spheroid Invasions") + +#### Drug Treatments + +![CN03](figures/CN03.png "CN03 Drug Treatment") + +![Y27632](figures/Y27632.png "Y27632 Drug Treatment") + +### Latent Dimensions + +![Model Training per Dimension](figures/SI_model_training.png "Model Training") + +## Usage + +Example notebooks can be found in [notebooks/](notebooks/). Source code can be found in [code/](code/). Briefly, the [Tensorflow](https://www.tensorflow.org/) implementation is found in [code/sunlab/sunflow/](code/sunlab/sunflow) and the [PyTorch](https://pytorch.org/) implementation can be found in [code/sunlab/sunflow/](code/sunlab/suntorch). Environments used can be found in the source Yaml files ready to be used with [Anaconda](https://www.anaconda.com/) or related technology. + +## Training + +An example of training a standard autoencoder can be found in [notebooks/Autoencoder.ipynb](notebooks/Autoencoder.ipynb). + +TODO: More implementations + +## Pretrained Model Information + +The MaxAbsScaler contains the scaling factors to transform the morphological features to the normalized features. The morphological features were derived from 1024x1024 pixel images on a confocal microscope (0.4NA, 10x objective) with a pixel to micron ratio of ??. -- cgit v1.2.1