version 1.0.2 release

Author: mrland99

build/lib/paste/PASTE.py

@@ -33,7 +33,7 @@ def pairwise_align(layer1, layer2, alpha = 0.1):
     pi, logw = ot.gromov.fused_gromov_wasserstein(M, D1.to_numpy(), D2.to_numpy(), a, b, loss_fun='square_loss', alpha= alpha, verbose=False, log=True)
     return pi
 
-def center_align(A, layers, lmbda, alpha = 0.1, n_components = 15, threshold = 0.001):
+def center_align(A, layers, lmbda, alpha = 0.1, n_components = 15, threshold = 0.001, random_seed = None):
     """
     Computes center alignment of layers.
     
@@ -61,7 +61,7 @@ def center_align(A, layers, lmbda, alpha = 0.1, n_components = 15, threshold = 0
     D = []
     for layer in layers:
         D.append(generateDistanceMatrix(layer, layer))
-    model = NMF(n_components=n_components, init='random', random_state=0)
+    model = NMF(n_components=n_components, solver = 'mu', beta_loss = 'kullback-leibler', init='random', random_state = random_seed)
     W = model.fit_transform(A.gene_exp)
     H = model.components_
     center_coordinates = A.coordinates
@@ -78,7 +78,7 @@ def center_align(A, layers, lmbda, alpha = 0.1, n_components = 15, threshold = 0
     while R_diff > 0.001 and iteration_count < 10:
         print("Iteration: " + str(iteration_count))
         pi, r = center_ot(W, H, layers, center_coordinates, common_genes, alpha)
-        W, H = center_NMF(W, H, layers, pi, lmbda, n_components)
+        W, H = center_NMF(W, H, layers, pi, lmbda, n_components, random_seed)
         R_new = np.dot(r,lmbda)
         iteration_count += 1
         R_diff = abs(R - R_new)
@@ -100,10 +100,10 @@ def center_ot(W, H, layers, center_coordinates, common_genes, alpha):
         r.append(r_q)
     return pi, np.array(r)
 
-def center_NMF(W, H, layers, pi, lmbda, n_components):
+def center_NMF(W, H, layers, pi, lmbda, n_components, random_seed):
     n = W.shape[0]
     B = n*sum([lmbda[i]*np.dot(pi[i], layers[i].gene_exp) for i in range(len(layers))])
-    model = NMF(n_components=n_components, init='random', random_state=0)
+    model = NMF(n_components=n_components, solver = 'mu', beta_loss = 'kullback-leibler', init='random', random_state = random_seed)
     W_new = model.fit_transform(B)
     H_new = model.components_
     return W_new, H_new

dist/paste-bio-1.0.1.tar.gz

@@ -1,6 +1,6 @@
 [metadata]
 name = paste-bio
-version = 1.0.1
+version = 1.0.2
 author = Max Land
 author_email = [email protected]
 description = A computational method to align and integrate spatial transcriptomics experiments.

dist/paste-bio-1.0.2.tar.gz

@@ -1,90 +1,92 @@
 Metadata-Version: 2.1
 Name: paste-bio
-Version: 1.0.1
+Version: 1.0.2
 Summary: A computational method to align and integrate spatial transcriptomics experiments.
 Home-page: https://github.com/raphael-group/paste
 Author: Max Land
 Author-email: [email protected]
 License: UNKNOWN
 Project-URL: Bug Tracker, https://github.com/raphael-group/paste/issues
-Description: # PASTE
-        
-        PASTE is a computational method that leverages both gene expression similarity and spatial distances between spots align and integrate spatial transcriptomics data. In particular, there are two methods:
-        1. `pairwise_align`: align spots across pairwise ST layers.
-        2. `center_align`: integrate multiple ST layers into one center layer.
-        
-        You can read our preprint [here](https://www.biorxiv.org/content/10.1101/2021.03.16.435604v1). 
-        
-        PASTE is actively being worked on with future updates coming. 
-        
-        ### Dependencies
-        
-        To run PASTE, you will need the following Python packages:
-        1. POT: Python Optimal Transport (https://PythonOT.github.io/)
-        2. NetworkX (https://networkx.org/)
-        3. Numpy
-        4. Pandas 
-        5. scipy.spatial
-        6. sklearn.preprocessing
-        
-        ### Installation
-        
-        The easiest way is to install PASTE on pypi: https://pypi.org/project/paste-bio/.
-        
-        `pip install paste-bio`
-        
-        Check out Tutorial.ipynb for an example of how to use PASTE.
-        
-        Or you can clone the respository and run from command line (see below).
-        
-        
-        ### Command Line
-        
-        We provide the option of running PASTE from the command line. 
-        
-        First, clone the repository:
-        
-        `git clone https://github.com/raphael-group/paste.git`
-        
-        Sample execution: `python paste-cmd-line.py -m pairwise -f file1.csv file2.csv file3.csv`
-        
-        Note: `pairwise` will return pairwise alignment between each consecutive pair of files (e.g. \[file1,file2\], \[file2,file3\]).
-        
-        | Flag | Name | Description | Default Value |
-        | --- | --- | --- | --- |
-        | -m | mode | Select either `pairwise` or `center` | (str) `pairwise` |
-        | -f | files | Path to data files (.csv) | None |
-        | -d | direc | Directory to store output files | Current Directory |
-        | -a | alpha | alpha parameter for PASTE | (float) `0.1` |
-        | -p | n_components | n_components for NMF step in `center_align` | (int) `15` |
-        | -l | lmbda | lambda parameter in `center_align` | (floats) probability vector of length `n`  |
-        | -i | intial_layer | Specify which file is also the intial layer in `center_align` | (int) `1` |
-        | -t | threshold | Convergence threshold for `center_align` | (float) `0.001` |
-        
-        Input files are .csv files of the form:
-        
-        ```
-               	'gene_a'  'gene_b'
-        '2x5'	   0         9      
-        '2x7'	   2         6      
-        ```
-        Where the columns indexes are gene names (str), row indexes are spatial coordinates (str), and entries are gene counts (int). In particular, row indexes are of the form `AxB` where `A` and `B` are floats.
-        
-        `pairwise_align` outputs a (.csv) file containing mapping of spots between each consecutive pair of layers. The rows correspond to spots of the first layer, and cols the second.
-        
-        `center_align` outputs two files containing the low dimensional representation (NMF decomposition) of the center layer gene expression, and files containing a mapping of spots between the center layer (rows) to each input layer (cols).
-        
-        ### Sample Dataset
-        
-        Added sample spatial transcriptomics dataset consisting of four breast cancer layers courtesy of:
-        
-        Ståhl, Patrik & Salmén, Fredrik & Vickovic, Sanja & Lundmark, Anna & Fernandez Navarro, Jose & Magnusson, Jens & Giacomello, Stefania & Asp, Michaela & Westholm, Jakub & Huss, Mikael & Mollbrink, Annelie & Linnarsson, Sten & Codeluppi, Simone & Borg, Åke & Pontén, Fredrik & Costea, Paul & Sahlén, Pelin Akan & Mulder, Jan & Bergmann, Olaf & Frisén, Jonas. (2016). Visualization and analysis of gene expression in tissue sections by spatial transcriptomics. Science. 353. 78-82. 10.1126/science.aaf2403. 
-        
-        Note: Original data is (.tsv), but we converted it to (.csv).
-        
 Platform: UNKNOWN
 Classifier: Programming Language :: Python :: 3
 Classifier: License :: OSI Approved :: BSD License
 Classifier: Operating System :: OS Independent
 Requires-Python: >=3.6
 Description-Content-Type: text/markdown
+License-File: LICENSE
+
+# PASTE
+
+PASTE is a computational method that leverages both gene expression similarity and spatial distances between spots align and integrate spatial transcriptomics data. In particular, there are two methods:
+1. `pairwise_align`: align spots across pairwise ST layers.
+2. `center_align`: integrate multiple ST layers into one center layer.
+
+You can read our preprint [here](https://www.biorxiv.org/content/10.1101/2021.03.16.435604v1). 
+
+PASTE is actively being worked on with future updates coming. 
+
+### Dependencies
+
+To run PASTE, you will need the following Python packages:
+1. POT: Python Optimal Transport (https://PythonOT.github.io/)
+3. Numpy
+4. Pandas 
+5. scipy.spatial
+6. sklearn.preprocessing
+
+### Installation
+
+The easiest way is to install PASTE on pypi: https://pypi.org/project/paste-bio/.
+
+`pip install paste-bio`
+
+Check out Tutorial.ipynb for an example of how to use PASTE.
+
+Or you can clone the respository and run from command line (see below).
+
+
+### Command Line
+
+We provide the option of running PASTE from the command line. 
+
+First, clone the repository:
+
+`git clone https://github.com/raphael-group/paste.git`
+
+Sample execution: `python paste-cmd-line.py -m pairwise -f file1.csv file2.csv file3.csv`
+
+Note: `pairwise` will return pairwise alignment between each consecutive pair of files (e.g. \[file1,file2\], \[file2,file3\]).
+
+| Flag | Name | Description | Default Value |
+| --- | --- | --- | --- |
+| -m | mode | Select either `pairwise` or `center` | (str) `pairwise` |
+| -f | files | Path to data files (.csv) | None |
+| -d | direc | Directory to store output files | Current Directory |
+| -a | alpha | alpha parameter for PASTE | (float) `0.1` |
+| -p | n_components | n_components for NMF step in `center_align` | (int) `15` |
+| -l | lmbda | lambda parameter in `center_align` | (floats) probability vector of length `n`  |
+| -i | intial_layer | Specify which file is also the intial layer in `center_align` | (int) `1` |
+| -t | threshold | Convergence threshold for `center_align` | (float) `0.001` |
+
+Input files are .csv files of the form:
+
+```
+       	'gene_a'  'gene_b'
+'2x5'	   0         9      
+'2x7'	   2         6      
+```
+Where the columns indexes are gene names (str), row indexes are spatial coordinates (str), and entries are gene counts (int). In particular, row indexes are of the form `AxB` where `A` and `B` are floats.
+
+`pairwise_align` outputs a (.csv) file containing mapping of spots between each consecutive pair of layers. The rows correspond to spots of the first layer, and cols the second.
+
+`center_align` outputs two files containing the low dimensional representation (NMF decomposition) of the center layer gene expression, and files containing a mapping of spots between the center layer (rows) to each input layer (cols).
+
+### Sample Dataset
+
+Added sample spatial transcriptomics dataset consisting of four breast cancer layers courtesy of:
+
+Ståhl, Patrik & Salmén, Fredrik & Vickovic, Sanja & Lundmark, Anna & Fernandez Navarro, Jose & Magnusson, Jens & Giacomello, Stefania & Asp, Michaela & Westholm, Jakub & Huss, Mikael & Mollbrink, Annelie & Linnarsson, Sten & Codeluppi, Simone & Borg, Åke & Pontén, Fredrik & Costea, Paul & Sahlén, Pelin Akan & Mulder, Jan & Bergmann, Olaf & Frisén, Jonas. (2016). Visualization and analysis of gene expression in tissue sections by spatial transcriptomics. Science. 353. 78-82. 10.1126/science.aaf2403. 
+
+Note: Original data is (.tsv), but we converted it to (.csv).
+
+