{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
    "\n",
    "import scanpy as sc\n",
    "import anndata as ad\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import anndata2ri\n",
    "import gdown\n",
    "import scipy\n",
    "import scipy.io\n",
    "from rpy2.robjects import r\n",
    "\n",
    "anndata2ri.activate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext rpy2.ipython"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "%%R\n",
    "suppressPackageStartupMessages({\n",
    "    library(SingleCellExperiment)\n",
    "    library(Seurat)\n",
    "    library(SeuratDisk)\n",
    "    library(Signac)\n",
    "})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Expressions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Download the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading...\n",
      "From: https://drive.google.com/uc?id=1E3AT2wztvswn3p_N5kWEmk0_ZHaT3OWH\n",
      "To: /mnt/home/icb/anastasia.litinetskaya/multigrate/example/chen2019-expressions.h5ad\n",
      "9.00MB [00:00, 83.4MB/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'chen2019-expressions.h5ad'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gdown.download(\n",
    "    'https://drive.google.com/uc?id=1E3AT2wztvswn3p_N5kWEmk0_ZHaT3OWH',\n",
    "    'chen2019-expressions.h5ad'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "adata = sc.read('chen2019-expressions.h5ad')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "adata_ = ad.AnnData(adata.layers['count'])\n",
    "adata_.obs_names = adata.obs_names\n",
    "adata_.var_names = adata.var_names\n",
    "adata_.obs['cell_type'] = adata.obs['cell_type']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "An object of class Seurat \n",
      "4000 features across 4793 samples within 1 assay \n",
      "Active assay: RNA (4000 features, 0 variable features)\n"
     ]
    }
   ],
   "source": [
    "%%R -i adata_\n",
    "rna = as.Seurat(adata_, counts='X', data=NULL)\n",
    "rna"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Peaks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Download the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading...\n",
      "From: https://drive.google.com/uc?id=13SdXwNDjG_UbnpCUnYqc2iHLwXEAjxNc\n",
      "To: /mnt/home/icb/anastasia.litinetskaya/multigrate/example/chen2019-peaks.h5ad\n",
      "106MB [00:01, 65.1MB/s] \n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'chen2019-peaks.h5ad'"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gdown.download(\n",
    "    'https://drive.google.com/uc?id=13SdXwNDjG_UbnpCUnYqc2iHLwXEAjxNc',\n",
    "    'chen2019-peaks.h5ad'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "adata = sc.read('chen2019-peaks.h5ad')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "adata_ = ad.AnnData(adata.layers['binary'])\n",
    "adata_.obs_names = adata.obs_names\n",
    "adata_.var_names = adata.var_names\n",
    "adata_.obs['cell_type'] = adata.obs['cell_type']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "An object of class Seurat \n",
      "40623 features across 4793 samples within 1 assay \n",
      "Active assay: RNA (40623 features, 0 variable features)\n"
     ]
    }
   ],
   "source": [
    "%%R -i adata_\n",
    "peaks = as.Seurat(adata_, counts='X', data=NULL)\n",
    "peaks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Run Seurat v4 WNN\n",
    "Following https://satijalab.org/seurat/v4.0/weighted_nearest_neighbor_analysis.html."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"Preprocessing RNA...\"\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "R[write to console]: Warning:\n",
      "R[write to console]:  The default method for RunUMAP has changed from calling Python UMAP via reticulate to the R-native UWOT using the cosine metric\n",
      "To use Python UMAP via reticulate, set umap.method to 'umap-learn' and metric to 'correlation'\n",
      "This message will be shown once per session\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1] \"Preprocessing ATAC...\"\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "R[write to console]: Calculating cell-specific modality weights\n",
      "\n",
      "R[write to console]: Finding 20 nearest neighbors for each modality.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  |                                                  | 0 % ~calculating   |+++++++++++++++++++++++++                         | 50% ~02s           |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=04s  \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "R[write to console]: Calculating kernel bandwidths\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  |                                                  | 0 % ~calculating   |+++++++++++++++++++++++++                         | 50% ~00s           |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=01s  \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "R[write to console]: Finding multimodal neighbors\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  |                                                  | 0 % ~calculating   |+++++++++++++++++++++++++                         | 50% ~06s           |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=13s  \n",
      "  |                                                  | 0 % ~calculating   |+++++++++++++++++++++++++                         | 50% ~01s           |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=01s  \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "R[write to console]: Constructing multimodal KNN graph\n",
      "\n",
      "R[write to console]: Constructing multimodal SNN graph\n",
      "\n",
      "R[write to console]: Computing sPCA transformation\n",
      "\n",
      "R[write to console]: Creating h5Seurat file for version 3.1.5.9900\n",
      "\n",
      "R[write to console]: Adding counts for RNA\n",
      "\n",
      "R[write to console]: Adding data for RNA\n",
      "\n",
      "R[write to console]: No variable features found for RNA\n",
      "\n",
      "R[write to console]: No feature-level metadata found for RNA\n",
      "\n",
      "R[write to console]: Adding counts for ATAC\n",
      "\n",
      "R[write to console]: Adding data for ATAC\n",
      "\n",
      "R[write to console]: Adding variable features for ATAC\n",
      "\n",
      "R[write to console]: Adding feature-level metadata for ATAC\n",
      "\n",
      "R[write to console]: Adding counts for SCT\n",
      "\n",
      "R[write to console]: Adding data for SCT\n",
      "\n",
      "R[write to console]: Adding scale.data for SCT\n",
      "\n",
      "R[write to console]: Adding variable features for SCT\n",
      "\n",
      "R[write to console]: No feature-level metadata found for SCT\n",
      "\n",
      "R[write to console]: Writing out SCTModel.list for SCT\n",
      "\n",
      "R[write to console]: Adding cell embeddings for pca\n",
      "\n",
      "R[write to console]: Adding loadings for pca\n",
      "\n",
      "R[write to console]: No projected loadings for pca\n",
      "\n",
      "R[write to console]: Adding standard deviations for pca\n",
      "\n",
      "R[write to console]: No JackStraw data for pca\n",
      "\n",
      "R[write to console]: Adding cell embeddings for umap.rna\n",
      "\n",
      "R[write to console]: No loadings for umap.rna\n",
      "\n",
      "R[write to console]: No projected loadings for umap.rna\n",
      "\n",
      "R[write to console]: No standard deviations for umap.rna\n",
      "\n",
      "R[write to console]: No JackStraw data for umap.rna\n",
      "\n",
      "R[write to console]: Adding cell embeddings for lsi\n",
      "\n",
      "R[write to console]: Adding loadings for lsi\n",
      "\n",
      "R[write to console]: No projected loadings for lsi\n",
      "\n",
      "R[write to console]: Adding standard deviations for lsi\n",
      "\n",
      "R[write to console]: No JackStraw data for lsi\n",
      "\n",
      "R[write to console]: Adding cell embeddings for umap.atac\n",
      "\n",
      "R[write to console]: No loadings for umap.atac\n",
      "\n",
      "R[write to console]: No projected loadings for umap.atac\n",
      "\n",
      "R[write to console]: No standard deviations for umap.atac\n",
      "\n",
      "R[write to console]: No JackStraw data for umap.atac\n",
      "\n",
      "R[write to console]: Adding cell embeddings for spca\n",
      "\n",
      "R[write to console]: Adding loadings for spca\n",
      "\n",
      "R[write to console]: No projected loadings for spca\n",
      "\n",
      "R[write to console]: Adding standard deviations for spca\n",
      "\n",
      "R[write to console]: No JackStraw data for spca\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%%R\n",
    "bm <- rna\n",
    "bm[[\"ATAC\"]] <- CreateAssayObject(counts = peaks@assays$RNA@counts)\n",
    "\n",
    "rm(rna)\n",
    "rm(peaks)\n",
    "\n",
    "print('Preprocessing RNA...')\n",
    "DefaultAssay(bm) <- \"RNA\"\n",
    "bm <- SCTransform(bm, variable.features.n = 4000, verbose=FALSE)\n",
    "bm <- RunPCA(bm, verbose=FALSE)\n",
    "bm <- RunUMAP(bm, dims = 1:50, reduction.name = 'umap.rna', reduction.key = 'rnaUMAP_', verbose=FALSE)\n",
    "\n",
    "print('Preprocessing ATAC...')\n",
    "DefaultAssay(bm) <- \"ATAC\"\n",
    "bm <- RunTFIDF(bm, verbose=FALSE)\n",
    "bm <- FindTopFeatures(bm, min.cutoff = 'q0', verbose=FALSE)\n",
    "bm <- RunSVD(bm, verbose=FALSE)\n",
    "bm <- RunUMAP(bm, reduction = 'lsi', dims = 2:50, reduction.name = \"umap.atac\", \n",
    "              reduction.key = \"atacUMAP_\", verbose=FALSE)\n",
    "\n",
    "bm <- FindMultiModalNeighbors(bm, reduction.list = list(\"pca\", \"lsi\"), \n",
    "                              dims.list = list(1:50, 2:50), modality.weight.name = \"RNA.weight\")\n",
    "\n",
    "bm <- RunSPCA(bm, assay = 'SCT', graph = 'wsnn', npcs = 20)\n",
    "\n",
    "SaveH5Seurat(bm, 'chen2019-seurat.h5seurat', overwrite = TRUE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "R[write to console]: Validating h5Seurat file\n",
      "\n",
      "R[write to console]: Adding scale.data from SCT as X\n",
      "\n",
      "R[write to console]: Adding data from SCT as raw\n",
      "\n",
      "R[write to console]: Transfering meta.data to obs\n",
      "\n",
      "R[write to console]: Adding dimensional reduction information for pca\n",
      "\n",
      "R[write to console]: Adding feature loadings for pca\n",
      "\n",
      "R[write to console]: Adding dimensional reduction information for spca\n",
      "\n",
      "R[write to console]: Adding feature loadings for spca\n",
      "\n",
      "R[write to console]: Adding dimensional reduction information for umap.rna\n",
      "\n",
      "R[write to console]: Adding dimensional reduction information for umap.atac (global)\n",
      "\n",
      "R[write to console]: Adding wsnn as neighbors\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%%R\n",
    "# wsnn graph is stored in SCT assay\n",
    "Convert(\"chen2019-seurat.h5seurat\", assay=\"SCT\", \"chen2019-seurat.h5ad\", overwrite=TRUE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AnnData object with n_obs × n_vars = 4793 × 3809\n",
       "    obs: 'cell_type', 'nCount_ATAC', 'nFeature_ATAC', 'nCount_SCT', 'nFeature_SCT', 'SCT.weight', 'ATAC.weight'\n",
       "    var: 'features'\n",
       "    uns: 'neighbors'\n",
       "    obsm: 'X_pca', 'X_spca', 'X_umap.atac', 'X_umap.rna'\n",
       "    varm: 'PCs', 'SPCA'\n",
       "    obsp: 'distances'"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "integrated = sc.read('chen2019-seurat.h5ad')\n",
    "integrated"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "latent = sc.AnnData(integrated.obsm['X_spca'])\n",
    "latent.obs = integrated.obs\n",
    "latent.obsp['connectivities'] = integrated.uns['neighbors']['distances']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "latent.write('../data/integrated/seurat/chen-seurat.h5ad')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "!rm 'chen2019-expressions.h5ad'\n",
    "!rm 'chen2019-peaks.h5ad'\n",
    "!rm 'chen2019-seurat.h5seurat'\n",
    "!rm 'chen2019-seurat.h5ad'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R version 4.0.3 (2020-10-10)\n",
      "Platform: x86_64-conda-linux-gnu (64-bit)\n",
      "Running under: CentOS Linux 7 (Core)\n",
      "\n",
      "Matrix products: default\n",
      "BLAS/LAPACK: /mnt/home/icb/anastasia.litinetskaya/miniconda3/envs/signac/lib/libopenblasp-r0.3.10.so\n",
      "\n",
      "locale:\n",
      " [1] LC_CTYPE=C                 LC_NUMERIC=C              \n",
      " [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    \n",
      " [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   \n",
      " [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 \n",
      " [9] LC_ADDRESS=C               LC_TELEPHONE=C            \n",
      "[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       \n",
      "\n",
      "attached base packages:\n",
      " [1] parallel  stats4    tools     stats     graphics  grDevices utils    \n",
      " [8] datasets  methods   base     \n",
      "\n",
      "other attached packages:\n",
      " [1] Matrix_1.3-2                Signac_1.1.1               \n",
      " [3] SeuratDisk_0.0.0.9018       SeuratObject_4.0.0         \n",
      " [5] Seurat_4.0.0                SingleCellExperiment_1.12.0\n",
      " [7] SummarizedExperiment_1.20.0 Biobase_2.50.0             \n",
      " [9] GenomicRanges_1.42.0        GenomeInfoDb_1.26.0        \n",
      "[11] IRanges_2.24.0              S4Vectors_0.28.0           \n",
      "[13] BiocGenerics_0.36.0         MatrixGenerics_1.2.0       \n",
      "[15] matrixStats_0.58.0         \n",
      "\n",
      "loaded via a namespace (and not attached):\n",
      "  [1] utf8_1.1.4               reticulate_1.18          tidyselect_1.1.0        \n",
      "  [4] RSQLite_2.2.3            AnnotationDbi_1.52.0     htmlwidgets_1.5.3       \n",
      "  [7] grid_4.0.3               BiocParallel_1.24.0      Rtsne_0.15              \n",
      " [10] munsell_0.5.0            codetools_0.2-18         ica_1.0-2               \n",
      " [13] future_1.21.0            miniUI_0.1.1.1           withr_2.4.1             \n",
      " [16] colorspace_2.0-0         OrganismDbi_1.32.0       knitr_1.31              \n",
      " [19] rstudioapi_0.13          ROCR_1.0-11              tensor_1.5              \n",
      " [22] listenv_0.8.0            GenomeInfoDbData_1.2.4   polyclip_1.10-0         \n",
      " [25] farver_2.1.0             bit64_4.0.5              parallelly_1.23.0       \n",
      " [28] vctrs_0.3.6              generics_0.1.0           xfun_0.20               \n",
      " [31] biovizBase_1.38.0        BiocFileCache_1.14.0     lsa_0.73.2              \n",
      " [34] ggseqlogo_0.1            R6_2.5.0                 hdf5r_1.3.3             \n",
      " [37] AnnotationFilter_1.14.0  reshape_0.8.8            bitops_1.0-6            \n",
      " [40] spatstat.utils_2.0-0     cachem_1.0.4             DelayedArray_0.16.0     \n",
      " [43] assertthat_0.2.1         promises_1.2.0.1         scales_1.1.1            \n",
      " [46] nnet_7.3-15              gtable_0.3.0             globals_0.14.0          \n",
      " [49] goftest_1.2-2            ggbio_1.38.0             ensembldb_2.14.0        \n",
      " [52] rlang_0.4.10             RcppRoll_0.3.0           splines_4.0.3           \n",
      " [55] rtracklayer_1.50.0       lazyeval_0.2.2           dichromat_2.0-0         \n",
      " [58] checkmate_2.0.0          BiocManager_1.30.10      reshape2_1.4.4          \n",
      " [61] abind_1.4-5              GenomicFeatures_1.42.0   backports_1.2.1         \n",
      " [64] httpuv_1.5.5             Hmisc_4.5-0              RBGL_1.66.0             \n",
      " [67] ggplot2_3.3.3            ellipsis_0.3.1           RColorBrewer_1.1-2      \n",
      " [70] ggridges_0.5.3           Rcpp_1.0.6               plyr_1.8.6              \n",
      " [73] base64enc_0.1-3          progress_1.2.2           zlibbioc_1.36.0         \n",
      " [76] purrr_0.3.4              RCurl_1.98-1.2           prettyunits_1.1.1       \n",
      " [79] rpart_4.1-15             openssl_1.4.3            deldir_0.2-10           \n",
      " [82] pbapply_1.4-3            cowplot_1.1.1            zoo_1.8-8               \n",
      " [85] ggrepel_0.9.1            cluster_2.1.1            magrittr_2.0.1          \n",
      " [88] RSpectra_0.16-0          data.table_1.14.0        scattermore_0.7         \n",
      " [91] lmtest_0.9-38            RANN_2.6.1               SnowballC_0.7.0         \n",
      " [94] ProtGenerics_1.22.0      fitdistrplus_1.1-3       hms_1.0.0               \n",
      " [97] patchwork_1.1.1          mime_0.10                xtable_1.8-4            \n",
      "[100] XML_3.99-0.5             jpeg_0.1-8.1             gridExtra_2.3           \n",
      "[103] compiler_4.0.3           biomaRt_2.46.0           tibble_3.1.0            \n",
      "[106] KernSmooth_2.23-18       crayon_1.4.1             htmltools_0.5.1.1       \n",
      "[109] mgcv_1.8-34              later_1.1.0.1            Formula_1.2-4           \n",
      "[112] tidyr_1.1.3              DBI_1.1.1                tweenr_1.0.1            \n",
      "[115] dbplyr_2.1.0             MASS_7.3-53.1            rappdirs_0.3.3          \n",
      "[118] cli_2.3.1                igraph_1.2.6             pkgconfig_2.0.3         \n",
      "[121] GenomicAlignments_1.26.0 foreign_0.8-81           plotly_4.9.3            \n",
      "[124] xml2_1.3.2               XVector_0.30.0           stringr_1.4.0           \n",
      "[127] VariantAnnotation_1.36.0 digest_0.6.27            sctransform_0.3.2       \n",
      "[130] RcppAnnoy_0.0.18         graph_1.68.0             spatstat.data_2.0-0     \n",
      "[133] Biostrings_2.58.0        fastmatch_1.1-0          leiden_0.3.7            \n",
      "[136] htmlTable_2.1.0          uwot_0.1.10              curl_4.3                \n",
      "[139] shiny_1.6.0              Rsamtools_2.6.0          lifecycle_1.0.0         \n",
      "[142] nlme_3.1-152             jsonlite_1.7.2           viridisLite_0.3.0       \n",
      "[145] askpass_1.1              BSgenome_1.58.0          fansi_0.4.2             \n",
      "[148] pillar_1.5.0             GGally_2.1.0             lattice_0.20-41         \n",
      "[151] fastmap_1.1.0            httr_1.4.2               survival_3.2-7          \n",
      "[154] glue_1.4.2               spatstat_1.64-1          png_0.1-7               \n",
      "[157] bit_4.0.4                ggforce_0.3.2            stringi_1.5.3           \n",
      "[160] blob_1.2.1               latticeExtra_0.6-29      memoise_2.0.0           \n",
      "[163] dplyr_1.0.4              irlba_2.3.3              future.apply_1.7.0      \n"
     ]
    }
   ],
   "source": [
    "%%R\n",
    "sessionInfo()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualize results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(latent.obs_names != adata.obs_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "latent.obs['cell_type'] = adata.obs['cell_type']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "sc.pp.neighbors(latent)\n",
    "sc.tl.umap(latent)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sc.pl.umap(latent, color='cell_type', frameon=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}