|
| 1 | +--- |
| 2 | +title: "Python Integration with anndataR" |
| 3 | +output: rmarkdown::html_vignette |
| 4 | +vignette: > |
| 5 | + %\VignetteIndexEntry{Python Integration with anndataR} |
| 6 | + %\VignetteEngine{knitr::rmarkdown} |
| 7 | + %\VignetteEncoding{UTF-8} |
| 8 | +--- |
| 9 | + |
| 10 | +```{r, include = FALSE} |
| 11 | +knitr::opts_chunk$set( |
| 12 | + collapse = TRUE, |
| 13 | + comment = "#>" |
| 14 | +) |
| 15 | +``` |
| 16 | + |
| 17 | +## Introduction |
| 18 | + |
| 19 | +anndataR works with Python AnnData objects through reticulate. |
| 20 | +You can load Python objects, apply Python functions to them, and |
| 21 | +convert to Seurat or SingleCellExperiment objects without manual conversion steps. |
| 22 | + |
| 23 | +## Basic Integration with Scanpy |
| 24 | + |
| 25 | +Install required Python packages if needed: |
| 26 | + |
| 27 | +```{r python_setup, eval=FALSE} |
| 28 | +reticulate::py_install("scanpy") |
| 29 | +``` |
| 30 | + |
| 31 | +```{r load_libraries} |
| 32 | +library(anndataR) |
| 33 | +library(reticulate) |
| 34 | +sc <- import("scanpy") |
| 35 | +``` |
| 36 | + |
| 37 | +Load a dataset directly from scanpy: |
| 38 | + |
| 39 | +```{r load_python_data} |
| 40 | +adata <- sc$datasets$pbmc3k_processed() |
| 41 | +print(adata) |
| 42 | +``` |
| 43 | + |
| 44 | +Apply scanpy functions directly: |
| 45 | + |
| 46 | +```{r apply_python_functions} |
| 47 | +sc$pp$filter_cells(adata, min_genes = 200L) |
| 48 | +sc$pp$normalize_total(adata, target_sum = 1e4) |
| 49 | +sc$pp$log1p(adata) |
| 50 | +``` |
| 51 | + |
| 52 | +Convert to Seurat object: |
| 53 | + |
| 54 | +```{r convert_to_seurat} |
| 55 | +seurat_obj <- adata$as_Seurat() |
| 56 | +print(seurat_obj) |
| 57 | +``` |
| 58 | + |
| 59 | +Convert to SingleCellExperiment object: |
| 60 | + |
| 61 | +```{r convert_to_sce} |
| 62 | +sce_obj <- adata$as_SingleCellExperiment() |
| 63 | +print(sce_obj) |
| 64 | +``` |
| 65 | + |
| 66 | +## Multi-modal Data with MuData |
| 67 | + |
| 68 | +Install required Python packages if needed: |
| 69 | + |
| 70 | +```{r mudata_setup, eval=FALSE} |
| 71 | +reticulate::py_install("mudata") |
| 72 | +``` |
| 73 | + |
| 74 | +```{r load_mudata} |
| 75 | +md <- import("mudata") |
| 76 | +``` |
| 77 | + |
| 78 | +Load a MuData object from file: |
| 79 | + |
| 80 | +```{r load_mudata_example} |
| 81 | +url <- "https://github.com/gtca/h5xx-datasets/raw/b1177ac8877c89d8bb355b072164384b4e9cc81d/datasets/minipbcite.h5mu" |
| 82 | +path <- tempfile(fileext = ".h5mu") |
| 83 | +download.file(url, path) |
| 84 | +mdata <- md$read_h5mu(path) |
| 85 | +``` |
| 86 | + |
| 87 | +Access individual modalities and convert them: |
| 88 | + |
| 89 | +```{r access_modalities} |
| 90 | +rna_mod <- mdata$mod[["rna"]] |
| 91 | +
|
| 92 | +rna_seurat <- rna_mod$as_Seurat() |
| 93 | +print(rna_seurat) |
| 94 | +
|
| 95 | +rna_sce <- rna_mod$as_SingleCellExperiment() |
| 96 | +print(rna_sce) |
| 97 | +``` |
| 98 | + |
0 commit comments