From 5f997a99f5771c1e55e68a102eb6552e2ffc0d7e Mon Sep 17 00:00:00 2001 From: Sam Hokin Date: Thu, 29 Apr 2021 10:25:24 -0600 Subject: [PATCH] Update README.md Just some spelling and grammar fixes. :) --- README.md | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/README.md b/README.md index 955c531..7d09c31 100644 --- a/README.md +++ b/README.md @@ -1,6 +1,6 @@ # GCNG using graph convolutional neural network and spaital transcriptomics data to infer cell-cell interactions -# Title, GCNG: Graph convolutional networks for inferring cell-cell interactions +# GCNG: Graph convolutional networks for inferring cell-cell interactions # https://genomebiology.biomedcentral.com/articles/10.1186/s13059-020-02214-w ## date: Nov 1, 2019 @@ -11,31 +11,31 @@ GCNG for extracellular gene relationship inference. (A) GCNG model using spatial ># 2, Code environment ->>## Users need to install python and ‘spektral’, ‘Keras’ and ‘Tensorflow’ modules, and all ohther modules required by the code. We recommend Anaconda to do this. +>>## Users need to install python and ‘spektral’, ‘Keras’ and ‘Tensorflow’ modules, and all other modules required by the code. We recommend Anaconda to do this. Author's environment is python 3.6.3 in a Linux server which is now running Centos 6.5 as the underlying OS and Rocks 6.1.1 as the cluster management revision. -Please used the old version of spektral as "spektral"github suggests : https://github.com/danielegrattarola/spektral#tensorflow-1-and-keras +Please used the old version of spektral as "spektral" github suggests : https://github.com/danielegrattarola/spektral#tensorflow-1-and-keras ># 3, Example for running Users should first set the path as the downloaded folder. ->>## 3.1 Training and test data generation for ligand-receptor prediction +>>## 3.1 Training and test data generation for ligand-receptor prediction >>>### Usage: python data_generation_interaction_ten_fold.py -`data_generation_interaction_ten_fold.py` uses the spatial location data to generate normalized adjacent matrix of cells, and save it in `seqfish_plus` folder; also uses the expression data to generate expression matrix for ten fold cross validation, and save it in `rand_1_10fold` folder. +`data_generation_interaction_ten_fold.py` uses the spatial location data to generate normalized adjacent matrix of cells, and saves it in `seqfish_plus` folder; also uses the expression data to generate expression matrix for ten fold cross validation, and saves it in `rand_1_10fold` folder. >>## 3.2 Training and test model python gcn_LR2_LR_as_nega_big.py - `gcn_LR2_LR_as_nega_big.py` exocrine GNCG that uses normalized adjacent matrix to generate normalized laplacian matrix, and then uses laplacian matrix to train and test GCNG models in ten fold cross validation. + `gcn_LR2_LR_as_nega_big.py` exocrine GNCG that uses normalized adjacent matrix to generate normalized laplacian matrix, and then uses laplacian matrix to train and test GCNG models in ten-fold cross validation. ( python gcn_LR2_LR_as_nega_big_plus_autocrine.py - `gcn_LR2_LR_as_nega_big_plus_autocrine.py` autocrine plus GNCG that uses adjacent matrix plus diagonal matrix to generate laplacian matrix, and then uses laplacian matrix to train and test GCNG models in ten fold cross validation + `gcn_LR2_LR_as_nega_big_plus_autocrine.py` autocrine plus GNCG that uses adjacent matrix plus diagonal matrix to generate Laplacian matrix, and then uses Laplacian matrix to train and test GCNG models in ten-fold cross validation. ) @@ -43,7 +43,7 @@ Users should first set the path as the downloaded folder. python gcn_LR2_LR_as_nega_big_layer_predict_min.py - `gcn_LR2_LR_as_nega_big_layer_predict_min.py` tries to find the optimal model during the trainning, by monitoring the validation dataset's accuracy. + `gcn_LR2_LR_as_nega_big_layer_predict_min.py` tries to find the optimal model during the training, by monitoring the validation dataset's accuracy. >>## 3.4 get performance of optimal model for ten fold cross validation