From b1c6fcef89c2d8c416a09de4c5820ff42200839a Mon Sep 17 00:00:00 2001 From: jonathan-columbiau <92599387+jonathan-columbiau@users.noreply.github.com> Date: Fri, 23 Feb 2024 22:45:09 +0000 Subject: [PATCH] =?UTF-8?q?Deploying=20to=20gh-pages=20from=20@=20jonathan?= =?UTF-8?q?-columbiau/lionmap@124b71846d7fe05d2c9cb90e3cc3df889da3ffbe=20?= =?UTF-8?q?=F0=9F=9A=80?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- index.html | 34 ++++++++++++++++++++++++++++------ pkgdown.yml | 2 +- search.json | 2 +- 3 files changed, 30 insertions(+), 8 deletions(-) diff --git a/index.html b/index.html index c488735..7cae51c 100644 --- a/index.html +++ b/index.html @@ -65,14 +65,36 @@
-

Hello! Welcome to the Lionmap package :)

+
+

lionmap +

+

Hello! Welcome to the Lionmap package :)

This package is meant to provide efficient and robust cell type classification while being easy to use.

-

Optional features that might increase mapping accuracy include using a custom hierarchy that describes the relationship between cell types, and setting different confidence thresholds for a classification to proceed to the next level of the hierarchy.

+

Optional features that might increase accuracy include using a custom hierarchy that describes the relationship between cell types, and setting different confidence thresholds for a classification to proceed to the next level of the hierarchy.

An example using a sample dataset where we run the entire pipeline (finding marker genes, creating models, and classifying cells) in a few lines of code is shown below.

-

```{r, eval=FALSE} #load reference and query datasets data(“hierarchy”,“train_ex_data_bpcells”,“test_ex_data_bpcells”,“train_ex_metadata”) #find marker genes marker_genes = FindMarkerGenes(train_ex_data_bpcells, train_ex_metadata, tree = hierarchy, metadata_cluster_column = “seurat_annotations”, metadata_cell_label_column = “cell_label”) #create models models = GetModels(marker_genes, train_ex_data_bpcells, train_ex_metadata, tree = hierarchy, metadata_cluster_column = “seurat_annotations”, metadata_cell_label_column = “cell_label”) #classify cells using models classifications = Classify(test_ex_data_bpcells, models, hierarchy)

-

```

-

The package only works using BPCells matrices as gene expression input. BPCells is a package that enables fast gene expression transformations and IO operations while requiring little memory on your computing environment. For more info on BPCells, check out the package here.

- +
+

Example +

+
+library(lionmap)
+#load reference and query datasets
+data("hierarchy","train_ex_data_bpcells","test_ex_data_bpcells","train_ex_metadata")
+#find marker genes 
+marker_genes = FindMarkerGenes(train_ex_data_bpcells, train_ex_metadata, tree = hierarchy, metadata_cluster_column = "seurat_annotations", metadata_cell_label_column = "cell_label")
+#create models
+models = GetModels(marker_genes, train_ex_data_bpcells, train_ex_metadata, tree = hierarchy, metadata_cluster_column = "seurat_annotations", metadata_cell_label_column = "cell_label")
+#classify cells using models
+classifications = Classify(test_ex_data_bpcells, models, hierarchy)
+
+
+

Installation +

+

You can install the development version of lionmap from GitHub with:

+
+# install.packages("devtools")
+devtools::install_github("jonathan-columbiau/lionmap")
+
+