Merge pull request #123 from KrishnaswamyLab/dev

Update to MAGIC v1.2.0

.gitignore

@@ -19,6 +19,8 @@ python/*.dll
 python/*.egg-info
 python/magic/__pycache__
 python/magic/*.pyc
+python/tutorial_notebooks/.ipynb_checkpoints
+
 matlab/EMT.csv
 
 # Mac detritus

README.md

@@ -1,6 +1,7 @@
 Markov Affinity-based Graph Imputation of Cells (MAGIC)
 -------------------------------------------------------
 
+[![Latest PyPI version](https://img.shields.io/pypi/v/magic-impute.svg)](https://pypi.org/project/magic-impute/)
 [![Latest CRAN version](https://img.shields.io/cran/v/Rmagic.svg)](https://cran.r-project.org/package=Rmagic)
 [![Travis CI Build](https://api.travis-ci.com/KrishnaswamyLab/MAGIC.svg?branch=master)](https://travis-ci.com/KrishnaswamyLab/MAGIC)
 [![Read the Docs](https://img.shields.io/readthedocs/magic.svg)](https://magic.readthedocs.io/)
@@ -47,7 +48,7 @@ MAGIC has been implemented in Python, Matlab, and R.
 
 To install with `pip`, run the following from a terminal:
 
-    pip install --user git+git://github.com/KrishnaswamyLab/MAGIC.git#subdirectory=python
+    pip install --user magic-impute
 
 #### Installation from GitHub
 
@@ -63,15 +64,15 @@ To clone the repository and install manually, run the following from a terminal:
 
 The following code runs MAGIC on test data located in the MAGIC repository.
 
-		import magic
-		import pandas as pd
-		import matplotlib.pyplot as plt
-		X = pd.read_csv("MAGIC/data/test_data.csv")
-		magic_operator = magic.MAGIC()
-		X_magic = magic_operator.fit_transform(X, genes=['VIM', 'CDH1', 'ZEB1'])
-		plt.scatter(X_magic['VIM'], X_magic['CDH1'], c=X_magic['ZEB1'], s=1, cmap='inferno')
-		plt.show()
-		magic.plot.animate_magic(X, gene_x='VIM', gene_y='CDH1', gene_color='ZEB1', operator=magic_operator)
+    import magic
+    import pandas as pd
+    import matplotlib.pyplot as plt
+    X = pd.read_csv("MAGIC/data/test_data.csv")
+    magic_operator = magic.MAGIC()
+    X_magic = magic_operator.fit_transform(X, genes=['VIM', 'CDH1', 'ZEB1'])
+    plt.scatter(X_magic['VIM'], X_magic['CDH1'], c=X_magic['ZEB1'], s=1, cmap='inferno')
+    plt.show()
+    magic.plot.animate_magic(X, gene_x='VIM', gene_y='CDH1', gene_color='ZEB1', operator=magic_operator)
 
 #### Tutorials
 
@@ -106,7 +107,7 @@ In R, run this command to install MAGIC and all dependencies:
 In a terminal, run the following command to install the Python
 repository.
 
-    pip install --user git+git://github.com/KrishnaswamyLab/MAGIC.git#subdirectory=python
+    pip install --user magic-impute
 
 #### Installation from GitHub
 
@@ -124,16 +125,16 @@ To clone the repository and install manually, run the following from a terminal:
 
 After installing the package, MAGIC can be run by loading the library and calling `magic()`:
 
-		library(Rmagic)
-		library(ggplot2)
-		data(magic_testdata)
-		MAGIC_data <- magic(magic_testdata, genes=c("VIM", "CDH1", "ZEB1"))
-		ggplot(MAGIC_data) +
-		  geom_point(aes(x=VIM, y=CDH1, color=ZEB1))
+    library(Rmagic)
+    library(ggplot2)
+    data(magic_testdata)
+    MAGIC_data <- magic(magic_testdata, genes=c("VIM", "CDH1", "ZEB1"))
+    ggplot(MAGIC_data) +
+      geom_point(aes(x=VIM, y=CDH1, color=ZEB1))
 
 #### Tutorials
 
-For a working example, see the Rmarkdown tutorials at https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/bonemarrow_tutorial.md and https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/EMT_tutorial.md or in `R/inst/examples`.
+For a working example, see the Rmarkdown tutorials at <http://htmlpreview.github.io/?https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/bonemarrow_tutorial.html> and <http://htmlpreview.github.io/?https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/EMT_tutorial.hmtl> or in `Rmagic/inst/examples`.
 
 ## Help
 

Rmagic/R/magic.R

@@ -1,5 +1,5 @@
 #' Perform MAGIC on a data matrix
-#' 
+#'
 #' Markov Affinity-based Graph Imputation of Cells (MAGIC) is an
 #' algorithm for denoising and transcript recover of single cells
 #' applied to single-cell RNA sequencing data, as described in
@@ -18,7 +18,7 @@
 #' power to which the diffusion operator is powered
 #' sets the level of diffusion. If 'auto', t is selected according to the
 #' Procrustes disparity of the diffused data.'
-#' @param npca number of PCA components that should be used; default: 20.
+#' @param npca number of PCA components that should be used; default: 100.
 #' @param init magic object, optional
 #' object to use for initialization. Avoids recomputing
 #' intermediate steps if parameters are the same.
@@ -83,7 +83,7 @@ magic <- function(data,
                   seed=NULL) {
   # check installation
   if (!reticulate::py_module_available(module = "magic")) {
-    install.magic()
+    load_pymagic()
   }
   tryCatch(pymagic, error = function(e) load_pymagic())
   k <- as.integer(k)

Rmagic/R/utils.R

@@ -9,20 +9,20 @@ null_equal <- function(x, y) {
   }
 }
 
-load_pymagic <- function() {
-  pymagic <<- reticulate::import("magic", delay_load = TRUE)
+load_pymagic <- function(delay_load = FALSE) {
+  result <- try(pymagic <<- reticulate::import("magic", delay_load = delay_load))
+  if (methods::is(result, "try-error")) {
+    install.magic()
+  }
 }
 
-#' Install PHATE Python Package
+#' Install MAGIC Python Package
 #'
-#' Install PHATE Python package into a virtualenv or conda env.
+#' Install MAGIC Python package into a virtualenv or conda env.
 #'
 #' On Linux and OS X the "virtualenv" method will be used by default
 #' ("conda" will be used if virtualenv isn't available). On Windows,
 #' the "conda" method is always used.
-#' As of reticulate v1.7, this functionality is only available in the
-#' development version of reticulate, which can be installed using
-#' `devtools::install_github('rstudio/reticulate')`
 #'
 #' @param envname Name of environment to install packages into
 #' @param method Installation method. By default, "auto" automatically finds
@@ -38,14 +38,24 @@ load_pymagic <- function() {
 #' @export
 install.magic <- function(envname = "r-reticulate", method = "auto",
                           conda = "auto", pip=TRUE, ...) {
-  stop(paste0(
-      "Cannot install MAGIC, please install from a console with ",
-      "pip install --user git+git://github.com/KrishnaswamyLab/MAGIC.git#subdirectory=python"
+  message("Attempting to install MAGIC python package with reticulate")
+  tryCatch({
+    reticulate::py_install("magic-impute",
+      envname = envname, method = method,
+      conda = conda, pip=pip, ...
+    )
+  },
+  error = function(e) {
+    stop(paste0(
+      "Cannot locate MAGIC Python package, please install through pip ",
+      "(e.g. pip install magic-impute)."
     ))
+  }
+  )
 }
 
 pymagic <- NULL
 
 .onLoad <- function(libname, pkgname) {
-  load_pymagic()
+  load_pymagic(delay_load = TRUE)
 }

Rmagic/README.Rmd

@@ -15,6 +15,7 @@ knitr::opts_chunk$set(
 )
 ```
 
+[![Latest PyPI version](https://img.shields.io/pypi/v/magic-impute.svg)](https://pypi.org/project/magic-impute/)
 [![Latest CRAN version](https://img.shields.io/cran/v/Rmagic.svg)](https://cran.r-project.org/package=Rmagic)
 [![Travis CI Build](https://api.travis-ci.com/KrishnaswamyLab/MAGIC.svg?branch=master)](https://travis-ci.com/KrishnaswamyLab/MAGIC)
 [![Read the Docs](https://img.shields.io/readthedocs/magic.svg)](https://magic.readthedocs.io/)
@@ -37,8 +38,8 @@ Markov Affinity-based Graph Imputation of Cells (MAGIC) is an algorithm for deno
 * MAGIC can be performed on a variety of datasets
 * Here, we show the usage of MAGIC on a toy dataset
 * You can view further examples of MAGIC on real data in our notebooks under `inst/examples`:
-    * https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/EMT_tutorial.md
-    * https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/bonemarrow_tutorial.md
+    * http://htmlpreview.github.io/?https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/EMT_tutorial.html
+    * http://htmlpreview.github.io/?https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/bonemarrow_tutorial.html
 
 ### Installation
 
@@ -57,7 +58,7 @@ install.packages("Rmagic")
 In a terminal, run the following command to install the Python repository.
 
 ```{bash install_python_magic, eval=FALSE}
-pip install --user git+git://github.com/KrishnaswamyLab/MAGIC.git#subdirectory=python
+pip install --user magic-impute
 ```
 
 #### Installaton from source

Rmagic/README.md

@@ -3,6 +3,7 @@ Rmagic v1.0.0
 
 <!-- README.md is generated from README.Rmd. Please edit that file -->
 
+[![Latest PyPI version](https://img.shields.io/pypi/v/magic-impute.svg)](https://pypi.org/project/magic-impute/)
 [![Latest CRAN version](https://img.shields.io/cran/v/Rmagic.svg)](https://cran.r-project.org/package=Rmagic)
 [![Travis CI
 Build](https://api.travis-ci.com/KrishnaswamyLab/MAGIC.svg?branch=master)](https://travis-ci.com/KrishnaswamyLab/MAGIC)
@@ -36,8 +37,8 @@ Diffusion*, Cell
   - You can view further examples of MAGIC on real data in our notebooks
     under
         `inst/examples`:
-      - <https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/EMT_tutorial.md>
-      - <https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/bonemarrow_tutorial.md>
+      - <http://htmlpreview.github.io/?https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/EMT_tutorial.html>
+      - <http://htmlpreview.github.io/?https://github.com/KrishnaswamyLab/MAGIC/blob/master/Rmagic/inst/examples/bonemarrow_tutorial.html>
 
 ### Installation
 
@@ -59,7 +60,7 @@ In a terminal, run the following command to install the Python
 repository.
 
 ``` bash
-pip install --user git+git://github.com/KrishnaswamyLab/MAGIC.git#subdirectory=python
+pip install --user magic-impute
 ```
 
 #### Installaton from source

Rmagic/inst/examples/bonemarrow_tutorial.Rmd

@@ -1,11 +1,13 @@
 ---
 title: "Rmagic Bone Marrow Tutorial"
 output:
-  github_document: default
-toc: true
+  html_document:
+    df_print: paged
+    toc: yes
+    toc_depth: '3'
 ---
 
-<!-- README.md is generated from README.Rmd. Please edit that file -->
+<!-- bonemarrow_tutorial.md is generated from bonemarrow_tutorial.Rmd. Please edit that file -->
 
 ```{r setup, include=FALSE}
 knitr::opts_chunk$set(echo = TRUE)
@@ -34,7 +36,7 @@ if (!require(Rmagic)) devtools::install_github("KrishnaswamyLab/magic/Rmagic")
 In a terminal, run the following command to install the Python repository.
 
 ```{bash install_python_magic, eval=FALSE}
-pip install --user git+git://github.com/KrishnaswamyLab/MAGIC.git#subdirectory=python
+pip install --user magic-impute
 ```
 
 We'll install a couple more tools for this tutorial.
@@ -66,7 +68,7 @@ library(phateR)
 
 ### Loading data
 
-In this tutorial, we will analyse myeloid and erythroid cells in mouse bone marrow, as described in Paul et al., 2015. The example data is located in the PHATE Github repository and we can load it directly from the web.
+In this tutorial, we will analyse myeloid and erythroid cells in mouse bone marrow, as described in Paul et al., 2015. The example data is located in the PHATE Github repository and we can load it directly from the web. You can run this tutorial with your own data by downloading <https://raw.githubusercontent.com/KrishnaswamyLab/MAGIC/master/Rmagic/inst/examples/bonemarrow_tutorial.Rmd> and opening it in RStudio.
 
 ```{r load_data}
 # load data
@@ -75,6 +77,8 @@ bmmsc <- bmmsc[,2:ncol(bmmsc)]
 bmmsc[1:5,1:10]
 ```
 
+### Filtering data
+
 First, we need to remove lowly expressed genes and cells with small library size.
 
 ```{r}
@@ -93,6 +97,7 @@ keep_rows <- rowSums(bmmsc) > 1000
 bmmsc <- bmmsc[keep_rows,]
 ```
 
+### Normalizing data
 
 We should library size normalize and transform the data prior to MAGIC. Many people use a log transform, which requires adding a "pseudocount" to avoid log(0). We square root instead, which has a similar form but doesn't suffer from instabilities at zero.
 
@@ -114,29 +119,31 @@ We can plot the data before and after MAGIC to visualize the results.
 
 ```{r plot_raw}
 ggplot(bmmsc) +
-  geom_point(aes(Mpo, Klf1, colour=Ifitm1)) +
-  scale_colour_viridis(option="B")
+  geom_point(aes(Mpo, Klf1, color=Ifitm1)) +
+  scale_color_viridis(option="B")
 ggsave('BMMSC_data_R_before_magic.png', width=5, height=5)
 ```
 
 The data suffers from dropout to the point that we cannot infer anything about the gene-gene relationships.
 
 ```{r plot_magic}
 ggplot(bmmsc_MAGIC) +
-  geom_point(aes(Mpo, Klf1, colour=Ifitm1)) +
-  scale_colour_viridis(option="B")
+  geom_point(aes(Mpo, Klf1, color=Ifitm1)) +
+  scale_color_viridis(option="B")
 ```
 
 As you can see, the gene-gene relationships are much clearer after MAGIC. These relationships also match the biological progression we expect to see - Ifitm1 is a stem cell marker, Klf1 is an erythroid marker, and Mpo is a myeloid marker.
 
+### Rerunning MAGIC with new parameters
+
 The data is a little too smooth - we can decrease `t` from the automatic value to reduce the amount of diffusion. We pass the original result to the argument `init` to avoid recomputing intermediate steps.
 
 ```{r decrease_t}
 bmmsc_MAGIC <- magic(bmmsc, genes=c("Mpo", "Klf1", "Ifitm1"), 
                      t=4, init=bmmsc_MAGIC)
 ggplot(bmmsc_MAGIC) +
-  geom_point(aes(Mpo, Klf1, colour=Ifitm1)) +
-  scale_colour_viridis(option="B")
+  geom_point(aes(Mpo, Klf1, color=Ifitm1)) +
+  scale_color_viridis(option="B")
 ggsave('BMMSC_data_R_after_magic.png', width=5, height=5)
 ```