biopipen 0.27.2__py3-none-any.whl → 0.27.3__py3-none-any.whl

biopipen/__init__.py

@@ -1 +1 @@
-__version__ = "0.27.2"
+__version__ = "0.27.3"

biopipen/ns/scrna.py

@@ -201,6 +201,13 @@ class SeuratPreparing(Proc):
                 - scvi: Same as `scVIIntegration`.
             - <more>: See <https://satijalab.org/seurat/reference/integratelayers>
 
+        DoubletFinder (ns): Arguments to run [`DoubletFinder`](https://github.com/chris-mcginnis-ucsf/DoubletFinder).
+            See also <https://demultiplexing-doublet-detecting-docs.readthedocs.io/en/latest/DoubletFinder.html>.
+            To disable `DoubletFinder`, set `envs.DoubletFinder` to `None` or `False`; or set `pcs` to `0`.
+            - PCs (type=int): Number of PCs to use for 'doubletFinder' function.
+            - doublets (type=float): Number of expected doublets as a proportion of the pool size.
+            - pN (type=float): Number of doublets to simulate as a proportion of the pool size.
+
     Requires:
         r-seurat:
             - check: {{proc.lang}} <(echo "library(Seurat)")
@@ -227,6 +234,7 @@ class SeuratPreparing(Proc):
             "min_cells": 5,
         },
         "IntegrateLayers": {"method": "harmony"},
+        "DoubletFinder": {"PCs": 0, "pN": 0.25, "doublets": 0.075},
     }
     script = "file://../scripts/scrna/SeuratPreparing.R"
     plugin_opts = {

biopipen/scripts/scrna/MarkersFinder.R

@@ -120,7 +120,7 @@ expand_each <- function(name, case) {
                     pull(case$each) %>% na.omit() %>% unique() %>% as.vector()
             }
             for (each in eachs) {
-                by <- make.names(paste0(".", name, "_", case$each,"_", each))
+                by <- make.names(paste0("..", name, "_", case$each,"_", each))
                 srtobj@meta.data <<- srtobj@meta.data %>% mutate(
                     !!sym(by) := if_else(
                         !!sym(case$each) == each,
@@ -364,6 +364,16 @@ add_case_report <- function(info, sigmarkers, siggenes) {
     }
 }
 
+ensure_sobj <- function(expr, allow_empty) {
+    tryCatch({ expr }, error = function(e) {
+        if (allow_empty) {
+            log_warn("  Ignoring this case: {e$message}")
+            return(NULL)
+        } else {
+            stop(e)
+        }
+    })
+}
 
 do_case_findall <- function(casename) {
     # casename
@@ -382,10 +392,17 @@ do_case_findall <- function(casename) {
     # args$min.cells.group <- args$min.cells.group %||% 1
     # args$min.cells.feature <- args$min.cells.feature %||% 1
     # args$min.pct <- args$min.pct %||% 0
+    allow_empty = startsWith(case$group.by, "..")
     if (!is.null(case$subset)) {
-        args$object <- srtobj %>% filter(!!parse_expr(case$subset) & !is.na(!!sym(case$group.by)))
+        args$object <- ensure_sobj({
+            srtobj %>% filter(!!parse_expr(case$subset) & !is.na(!!sym(case$group.by)))
+        }, allow_empty)
+        if (is.null(args$object)) { return() }
     } else {
-        args$object <- srtobj %>% filter(!is.na(!!sym(case$group.by)))
+        args$object <- ensure_sobj({
+            srtobj %>% filter(!is.na(!!sym(case$group.by)))
+        }, allow_empty)
+        if (is.null(args$object)) { return() }
     }
     Idents(args$object) <- case$group.by
 
@@ -486,11 +503,19 @@ do_case <- function(casename) {
     # sigmarkers
     # rest
     args <- case$rest
+    allow_empty = startsWith(case$group.by, "..")
     if (!is.null(case$subset)) {
-        args$object <- srtobj %>% filter(!!parse_expr(case$subset) & !is.na(!!sym(case$group.by)))
+        args$object <- ensure_sobj({
+            srtobj %>% filter(!!parse_expr(case$subset) & !is.na(!!sym(case$group.by)))
+        }, allow_empty)
+        if (is.null(args$object)) { return() }
     } else {
-        args$object <- srtobj %>% filter(!is.na(!!sym(case$group.by)))
+        args$object <- ensure_sobj({
+            srtobj %>% filter(!is.na(!!sym(case$group.by)))
+        }, allow_empty)
+        if (is.null(args$object)) { return() }
     }
+
     args$assay <- case$assay
     args$group.by <- case$group.by
     args$ident.1 <- case$ident.1

biopipen/scripts/scrna/MetaMarkers.R

@@ -76,7 +76,7 @@ expand_each <- function(name, case) {
                 pull(case$each) %>% unique() %>% na.omit()
         }
         for (each in eachs) {
-            by = make.names(paste0(".", name, "_", case$each, "_", each))
+            by = make.names(paste0("..", name, "_", case$each, "_", each))
             idents <- case$idents
             if (is.null(idents) || length(idents) == 0) {
                 srtobj@meta.data = srtobj@meta.data %>%
@@ -169,17 +169,31 @@ do_enrich <- function(info, markers, sig) {
     }
 }
 
+ensure_sobj <- function(expr, allow_empty) {
+    tryCatch({ expr }, error = function(e) {
+        if (allow_empty) {
+            log_warn("  Ignoring this case: {e$message}")
+            return(NULL)
+        } else {
+            stop(e)
+        }
+    })
+}
+
 do_case <- function(casename) {
     log_info("- Dealing with case: {casename} ...")
     info <- casename_info(casename, cases, outdir, create = TRUE)
     case <- cases[[casename]]
+    allow_empty = startsWith(case$group_by, "..")
 
     if (sum(!is.na(srtobj@meta.data[[case$group_by]])) == 0) {
         msg = "Not enough cells to run tests."
     } else {
-        sobj <- srtobj %>% filter(!is.na(!!sym(case$group_by)))
+        sobj <- ensure_sobj({ srtobj %>% filter(!is.na(!!sym(case$group_by))) }, allow_empty)
+        if (is.null(sobj)) { return() }
         if (!is.null(case$subset)) {
-            sobj <- srtobj %>% filter(!is.na(!!sym(case$group_by)), !!parse_expr(case$subset))
+            sobj <- ensure_sobj({ sobj %>% filter(!!parse_expr(case$subset)) }, allow_empty)
+            if (is.null(sobj)) { return() }
         }
         df <- tryCatch({
                 GetAssayData(sobj, layer = "data")

biopipen/scripts/scrna/ScFGSEA.R

@@ -72,7 +72,7 @@ expand_each <- function(name, case) {
                 pull(case$each) %>% na.omit() %>% unique() %>% as.vector()
         }
         for (each in eachs) {
-            by <- make.names(paste0(".", name, "_", case$each,"_", each))
+            by <- make.names(paste0("..", name, "_", case$each,"_", each))
             srtobj@meta.data <<- srtobj@meta.data %>%
                 mutate(!!sym(by) := if_else(
                     !!sym(case$each) == each,
@@ -97,18 +97,35 @@ log_info("- Expanding cases...")
 cases <- expand_cases(cases, defaults, expand_each)
 
 
+ensure_sobj <- function(expr, allow_empty) {
+    tryCatch({ expr }, error = function(e) {
+        if (allow_empty) {
+            log_warn("  Ignoring this case: {e$message}")
+            return(NULL)
+        } else {
+            stop(e)
+        }
+    })
+}
+
+
 do_case <- function(name, case) {
     log_info("- Handling case: {name} ...")
     info <- casename_info(name, cases, outdir, create = TRUE)
 
+    allow_empty = startsWith(case$group.by, "..")
     # prepare expression matrix
     log_info("  Preparing expression matrix...")
-    sobj <- srtobj %>% filter(!is.na(!!sym(case$group.by)))
+    sobj <- ensure_sobj({ srtobj %>% filter(!is.na(!!sym(case$group.by))) }, allow_empty)
+    if (is.null(sobj)) { return() }
+
     if (!is.null(case$subset)) {
-        sobj <- sobj %>% filter(!!!parse_exprs(case$subset))
+        sobj <- ensure_sobj({ sobj %>% filter(!!!parse_exprs(case$subset)) }, allow_empty)
+        if (is.null(sobj)) { return() }
     }
     if (!is.null(case$ident.2)) {
-        sobj <- sobj %>% filter(!!sym(case$group.by) %in% c(case$ident.1, case$ident.2))
+        sobj <- ensure_sobj({ sobj %>% filter(!!sym(case$group.by) %in% c(case$ident.1, case$ident.2)) }, allow_empty)
+        if (is.null(sobj)) { return() }
     }
 
     allclasses <- sobj@meta.data[, case$group.by, drop = TRUE]

biopipen/scripts/scrna/SeuratPreparing.R

@@ -13,6 +13,7 @@ envs = {{envs | r: todot = "-", skip = 1}}
 
 set.seed(8525)
 options(future.globals.maxSize = 80000 * 1024^2)
+options(future.rng.onMisuse="ignore")
 options(Seurat.object.assay.version = "v5")
 plan(strategy = "multicore", workers = envs$ncores)
 
@@ -342,7 +343,7 @@ RunPCAArgs$object <- sobj
 sobj <- do_call(RunPCA, RunPCAArgs)
 
 if (!envs$no_integration) {
-    log_info("- Running IntegrateLayers ...")
+    log_info("- Running IntegrateLayers (method = {envs$IntegrateLayers$method}) ...")
     IntegrateLayersArgs <- envs$IntegrateLayers
     method <- IntegrateLayersArgs$method
     if (!is.null(IntegrateLayersArgs$reference) && is.character(IntegrateLayersArgs$reference)) {
@@ -383,6 +384,117 @@ if (!envs$use_sct) {
     sobj <- JoinLayers(sobj)
 }
 
+if (!is.null(envs$DoubletFinder) && is.list(envs$DoubletFinder) && envs$DoubletFinder$PCs > 0) {
+    library(DoubletFinder)
+
+    log_info("Running DoubletFinder ...")
+    log_info("- Preparing Seurat object ...")
+    # More controls from envs?
+    sobj <- FindNeighbors(sobj, dims = 1:envs$DoubletFinder$PCs)
+    sobj <- FindClusters(sobj)
+
+    log_info("- pK Indentification ...")
+    sweep.res.list <- paramSweep(
+        sobj,
+        PCs = 1:envs$DoubletFinder$PCs,
+        sct = envs$use_sct,
+        num.cores = envs$ncores
+    )
+    sweep.stats <- summarizeSweep(sweep.res.list, GT = FALSE)
+    bcmvn <- find.pK(sweep.stats)
+
+    bcmvn$Selected <- bcmvn$pK == bcmvn$pK[which.max(bcmvn$BCmetric)[1]]
+    plot <- ggplot(bcmvn, aes(x = pK, y = BCmetric, color = Selected)) +
+        geom_point() +
+        # rotate x axis labels
+        theme(axis.text.x = element_text(angle = 90, hjust = 1))
+    ggsave(plot, filename = file.path(plotsdir, "pK_BCmetric.png"))
+
+    pK <- bcmvn$pK[which.max(bcmvn$BCmetric)[1]]
+    pK <- as.numeric(as.character(pK))
+    pN <- envs$DoubletFinder$pN
+    log_info("- Homotypic Doublet Proportion Estimate ...")
+    homotypic.prop <- modelHomotypic(Idents(sobj))
+    nExp_poi <- round(nrow(sobj@meta.data) * envs$DoubletFinder$doublets)
+    nExp_poi.adj <- round(nExp_poi * (1 - homotypic.prop))
+
+    log_info("- Running DoubletFinder ...")
+    sobj <- doubletFinder(
+        sobj,
+        PCs = 1:envs$DoubletFinder$PCs,
+        pN = pN,
+        pK = pK,
+        nExp = nExp_poi.adj,
+        reuse.pANN = FALSE,
+        sct = envs$use_sct
+    )
+    pANN_col <- paste0("pANN_", pN, "_", pK)
+    pANN_col <- colnames(sobj@meta.data)[grepl(pANN_col, colnames(sobj@meta.data))]
+    DF_col <- paste0("DF.classifications_", pN, "_", pK)
+    DF_col <- colnames(sobj@meta.data)[grepl(DF_col, colnames(sobj@meta.data))]
+    doublets <- as.data.frame(
+        cbind(
+            colnames(sobj),
+            sobj@meta.data[, pANN_col],
+            sobj@meta.data[, DF_col]
+        )
+    )
+    colnames(doublets) <-  c("Barcode","DoubletFinder_score","DoubletFinder_DropletType")
+    write.table(
+        doublets,
+        file.path(joboutdir, "DoubletFinder_doublets_singlets.txt"),
+        row.names = FALSE,
+        quote = FALSE,
+        sep = "\t"
+    )
+
+    summary <- as.data.frame(table(doublets$DoubletFinder_DropletType))
+    colnames(summary) <- c("Classification", "Droplet_N")
+    write.table(
+        summary,
+        file.path(joboutdir, "DoubletFinder_summary.txt"),
+        row.names = FALSE,
+        quote = FALSE,
+        sep = "\t"
+    )
+
+    # Do a dimplot
+    log_info("- Plotting dimension reduction ...")
+    dimp <- DimPlot(
+        sobj, group.by = DF_col, order = "Doublet",
+        cols = c("#333333", "#FF3333"), pt.size = 0.8, alpha = 0.5)
+    ggsave(dimp, filename = file.path(plotsdir, "DoubletFinder_dimplot.png"))
+
+    log_info("- Filtering doublets ...")
+    sobj <- subset(sobj, cells = doublets$Barcode[doublets$DoubletFinder_DropletType == "Singlet"])
+
+    add_report(
+        list(
+            kind = "descr",
+            content = "The table contains the number of cells classified as singlets and doublets."
+        ),
+        list(
+            kind = "table",
+            data = list(path = file.path(joboutdir, "DoubletFinder_summary.txt"))
+        ),
+        h1 = "DoubletFinder Results",
+        h2 = "The DoubletFinder Summary"
+    )
+    add_report(
+        list(
+            name = "pK vs BCmetric",
+            src = file.path(plotsdir, "pK_BCmetric.png")
+        ),
+        list(
+            name = "Dimension Reduction Plot",
+            src = file.path(plotsdir, "DoubletFinder_dimplot.png")
+        ),
+        ui = "table_of_images",
+        h1 = "DoubletFinder Results",
+        h2 = "Plots"
+    )
+}
+
 log_info("Saving filtered seurat object ...")
 saveRDS(sobj, rdsfile)
 

{biopipen-0.27.2.dist-info → biopipen-0.27.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: biopipen
-Version: 0.27.2
+Version: 0.27.3
 Summary: Bioinformatics processes/pipelines that can be run from `pipen run`
 License: MIT
 Author: pwwang
@@ -17,6 +17,7 @@ Requires-Dist: datar[pandas] (>=0.15.6,<0.16.0)
 Requires-Dist: pipen-board[report] (>=0.15,<0.16)
 Requires-Dist: pipen-cli-run (>=0.13,<0.14)
 Requires-Dist: pipen-filters (>=0.12,<0.13)
-Requires-Dist: pipen-poplog (>=0.1,<0.2)
+Requires-Dist: pipen-poplog (>=0.1.2,<0.2.0)
 Requires-Dist: pipen-runinfo (>=0.6,<0.7) ; extra == "runinfo"
 Requires-Dist: pipen-verbose (>=0.11,<0.12)
+Requires-Dist: pyyaml-include (==1.*)

{biopipen-0.27.2.dist-info → biopipen-0.27.3.dist-info}/RECORD

@@ -1,4 +1,4 @@
-biopipen/__init__.py,sha256=_CerHyxnUJ2hU0sB7noT3JBjjI0ohEM5boAPBoFxOg4,23
+biopipen/__init__.py,sha256=lxhjPOOCzhlHB02EzaqTtDdBFZSOLV3WLWw2HC0DYvo,23
 biopipen/core/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 biopipen/core/config.py,sha256=edK5xnDhM8j27srDzsxubi934NMrglLoKrdcC8qsEPk,1069
 biopipen/core/config.toml,sha256=20RCI30Peee1EQdfb_UbV3Hf74XUPndJnYZlUThytsw,1781
@@ -21,7 +21,7 @@ biopipen/ns/gsea.py,sha256=EsNRAPYsagaV2KYgr4Jv0KCnZGqayM209v4yOGGTIOI,7423
 biopipen/ns/misc.py,sha256=fzn0pXvdghMkQhu-e3MMapPNMyO6IAJbtTzVU3GbFa0,3246
 biopipen/ns/plot.py,sha256=yguxmErUOH-hOM10JfuI_sXw2p49XF8yGR_gXfbd5yQ,4066
 biopipen/ns/rnaseq.py,sha256=bKAa6friFWof4yDTWZQahm1MS-lrdetO1GqDKdfxXYc,7708
-biopipen/ns/scrna.py,sha256=eLCXGyVcgq3vQ-br5SFHHmfIVPaJN4kSFxtCiJiYamg,102716
+biopipen/ns/scrna.py,sha256=i9h0xNOII3SqJ_cJOZ5epn8breAsc-yXH_Us04DoZvg,103401
 biopipen/ns/scrna_metabolic_landscape.py,sha256=9s1NvH3aMaNDXyfwy9TdzGcSP_lIW4JqhLgknNZcIKE,28313
 biopipen/ns/snp.py,sha256=Nq20NJzQ9YiqE9mhtCUH6dfs7528o1e4N-j9PewjAsQ,3016
 biopipen/ns/stats.py,sha256=yJ6C1CXF84T7DDs9mgufqUOr89Rl6kybE5ji8Vnx6cw,13693
@@ -129,12 +129,12 @@ biopipen/scripts/scrna/ExprImpution-alra.R,sha256=w3W1txJcdWg52-SETY2Z0lO7maDNfi
 biopipen/scripts/scrna/ExprImpution-rmagic.R,sha256=jYIfqZpnvjKJkvItLnemPVtUApHBYQi1_L8rHVbEe1M,735
 biopipen/scripts/scrna/ExprImpution-scimpute.R,sha256=mg40qCUW7-nP5oHPvARq7dmtoahM0GRFWXQpum0BXVk,1082
 biopipen/scripts/scrna/ExprImpution.R,sha256=7768ezrr59xUZDXq8lO9jj2XhnkSsx-xxBmOD9_DO7c,313
-biopipen/scripts/scrna/MarkersFinder.R,sha256=TvLVozCsgL_R-EMW7SbkCRdpchxt9k7Ewwz5nb3TOYo,22172
-biopipen/scripts/scrna/MetaMarkers.R,sha256=J__ZZ4K4P-Jdty1lZhRldu4rAErLxMtDZkRUlOqZea4,10852
+biopipen/scripts/scrna/MarkersFinder.R,sha256=M7fHTbHHErZ9JbLmjDkx-6yVIay0_h0MkvgFegnqL44,22918
+biopipen/scripts/scrna/MetaMarkers.R,sha256=9ve1X0TrDzS_ZEW6HtU3n8R-uPx7q-hYMMNFVDSE8wQ,11272
 biopipen/scripts/scrna/ModuleScoreCalculator.R,sha256=JSHd-_-KiFqW8avCGxgU4T-C5BtDr2u0kwIvEu2lFIg,4188
 biopipen/scripts/scrna/RadarPlots.R,sha256=TGPUTUcHOHgd9rsNtLYT-N6WHiFNDBZsiIoqkyAJh0A,13020
 biopipen/scripts/scrna/SCImpute.R,sha256=dSJOHhmJ3x_72LBRXT72dbCti5oiB85CJ-OjWtqONbk,2958
-biopipen/scripts/scrna/ScFGSEA.R,sha256=M6YeqUNa_0bq1qmL8dutQR3o5v2jy_gICCLaWw5c3A4,5738
+biopipen/scripts/scrna/ScFGSEA.R,sha256=2UCTCIydVkPGvn7WP-_fcE7857iKKDxY56-j-ruyO8o,6254
 biopipen/scripts/scrna/Seurat2AnnData.R,sha256=qz4u-B5J3GMwttubnNnByJXreziFbrP5Mak0L0q7eG0,1557
 biopipen/scripts/scrna/SeuratClusterStats-dimplots.R,sha256=gViDgQ8NorYD64iK0FgcODOrDOw0tExZmhuPRuLNp4g,2354
 biopipen/scripts/scrna/SeuratClusterStats-features.R,sha256=SaKTJloP1fttRXZQeb2ApX0ej7al13wOoEYkthSk13k,15489
@@ -147,7 +147,7 @@ biopipen/scripts/scrna/SeuratFilter.R,sha256=BrYK0MLdaTtQvInMaQsmOt7oH_hlks0M1zy
 biopipen/scripts/scrna/SeuratLoading.R,sha256=ekWKnHIqtQb3kHVQiVymAHXXqiUxs6KKefjZKjaykmk,900
 biopipen/scripts/scrna/SeuratMap2Ref.R,sha256=Xn3VnvKqShuC0Ju05380wjuLVSdW0uWVzntdxjme244,4359
 biopipen/scripts/scrna/SeuratMetadataMutater.R,sha256=Pp4GsF3hZ6ZC2vroC3LSBmVa4B1p2L3hbh981yaAIeQ,1093
-biopipen/scripts/scrna/SeuratPreparing.R,sha256=cgXWon2it6g4y-yrYk_zhivViOX8ZVf36u3wb9lKtj0,13133
+biopipen/scripts/scrna/SeuratPreparing.R,sha256=c_aBM0mugBNyYJ5OjNVDR_Cj0sGqkiJZXCOk3pesFDk,16990
 biopipen/scripts/scrna/SeuratSplit.R,sha256=vdK11V39_Uo_NaOh76QWCtxObGaEr5Ynxqq0hTiSvsU,754
 biopipen/scripts/scrna/SeuratSubClustering.R,sha256=L1SwKhNNKvsQGrcj0ZjScW9BLuvdO2pg7U48Ospsot8,6096
 biopipen/scripts/scrna/SeuratSubset.R,sha256=yVA11NVE2FSSw-DhxQcJRapns0tNNHdyDYi5epO6SKM,1776
@@ -238,7 +238,7 @@ biopipen/utils/reference.py,sha256=6bPSwQa-GiDfr7xLR9a5T64Ey40y24yn3QfQ5wDFZkU,4
 biopipen/utils/rnaseq.R,sha256=Ro2B2dG-Z2oVaT5tkwp9RHBz4dp_RF-JcizlM5GYXFs,1298
 biopipen/utils/single_cell.R,sha256=pJjYP8bIZpNAtTQ32rOXhZxaM1Y-6D-xUcK3pql9tbk,4316
 biopipen/utils/vcf.py,sha256=ajXs0M_QghEctlvUlSRjWQIABVF02wPdYd-0LP4mIsU,9377
-biopipen-0.27.2.dist-info/METADATA,sha256=G3Cij0e6iDpAZ87kUeVZabL8MHa689YLGzIFIjaLmG4,878
-biopipen-0.27.2.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-biopipen-0.27.2.dist-info/entry_points.txt,sha256=wu70aoBcv1UahVbB_5237MY-9M9_mzqmWjDD-oi3yz0,621
-biopipen-0.27.2.dist-info/RECORD,,
+biopipen-0.27.3.dist-info/METADATA,sha256=4DeAjhBZHdg7pZXoTNPiQkzGsx6hSm7VwgWgyYKMY18,920
+biopipen-0.27.3.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+biopipen-0.27.3.dist-info/entry_points.txt,sha256=wu70aoBcv1UahVbB_5237MY-9M9_mzqmWjDD-oi3yz0,621
+biopipen-0.27.3.dist-info/RECORD,,