biopipen 0.27.1__tar.gz → 0.27.3__tar.gz

{biopipen-0.27.1 → biopipen-0.27.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: biopipen
-Version: 0.27.1
+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-0.27.3/biopipen/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.27.3"

{biopipen-0.27.1 → biopipen-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-0.27.1 → biopipen-0.27.3}/biopipen/ns/snp.py

@@ -7,12 +7,15 @@ from ..core.config import config
 class PlinkSimulation(Proc):
     """Simulate SNPs using PLINK v1.9
 
-    See also <https://www.cog-genomics.org/plink/1.9/input#simulate>.
+    See also <https://www.cog-genomics.org/plink/1.9/input#simulate> and
+    <https://pwwang.github.io/biopipen/api/biopipen.ns.snp/#biopipen.ns.snp.PlinkSimulation>
 
     Input:
-        nsnps: Number of SNPs to simulate
-        ncases: Number of cases to simulate
-        nctrls: Number of controls to simulate
+        configfile: Configuration file containing the parameters for the simulation.
+            The configuration file (in toml, yaml or json format) should contain a
+            dictionary of parameters.  The parameters are listed in `envs` except
+            `ncores`, which is used for parallelization. You can set parameters
+            in `envs` and override them in the configuration file.
 
     Output:
         outdir: Output directory containing the simulated data
@@ -21,9 +24,11 @@ class PlinkSimulation(Proc):
             SNPs and columns representing samples.
 
     Envs:
+        nsnps (type=int): Number of SNPs to simulate
+        ncases (type=int): Number of cases to simulate
+        nctrls (type=int): Number of controls to simulate
         plink: Path to PLINK v1.9
-        seed (type=int): Random seed.
-            If not set, seed will not be set.
+        seed (type=int): Random seed. If not set, seed will not be set.
         label: Prefix label for the SNPs.
         prevalence  (type=float): Disease prevalence.
         minfreq (type=float): Minimum allele frequency.
@@ -41,19 +46,17 @@ class PlinkSimulation(Proc):
             This only affects the sample names in the genotype matrix file
             (`out.gtmat`).
     """
-    input = "nsnps:var, ncases:var, nctrls:var"
+    input = "configfile:file"
     output = [
-        (
-            "outdir:dir:{{in.nsnps | int}}_"
-            "{{in.ncases | int}}xcases_{{in.nctrls | int}}xctrls.plink_sim"
-        ),
-        (
-            "gtmat:file:{{in.nsnps | int}}_"
-            "{{in.ncases | int}}xcases_{{in.nctrls | int}}xctrls.plink_sim/gtmat.txt"
-        ),
+        "outdir:dir:{{in.configfile | stem}}.plink_sim",
+        "gtmat:file:{{in.configfile | stem}}.plink_sim/"
+        "{{in.configfile | stem}}-gtmat.txt",
     ]
     lang = config.lang.python
     envs = {
+        "nsnps": None,
+        "ncases": None,
+        "nctrls": None,
         "plink": config.exe.plink,
         "seed": None,
         "label": "SNP",

{biopipen-0.27.1 → biopipen-0.27.3}/biopipen/ns/tcr.py

@@ -983,6 +983,7 @@ class CloneResidency(Proc):
             before calculating the clone residency. For example, `Clones > 1` to filter
             out singletons.
         prefix: The prefix of the cell barcodes in the `Seurat` object.
+        upset_ymax: The maximum value of the y-axis in the upset bar plots.
         upset_trans: The transformation to apply to the y axis of upset bar plots.
             For example, `log10` or `sqrt`. If not specified, the y axis will be
             plotted as is. Note that the position of the bar plots will be dodged
@@ -1007,6 +1008,7 @@ class CloneResidency(Proc):
         "mutaters": {},
         "subset": None,
         "prefix": "{Sample}_",
+        "upset_ymax": None,
         "upset_trans": None,
         "cases": {},
     }
@@ -1595,3 +1597,74 @@ class TESSA(Proc):
     }
     script = "file://../scripts/tcr/TESSA.R"
     plugin_opts = {"report": "file://../reports/tcr/TESSA.svelte"}
+
+
+class TCRDock(Proc):
+    """Using TCRDock to predict the structure of MHC-peptide-TCR complexes
+
+    See <https://github.com/phbradley/TCRdock>.
+
+    Input:
+        configfile: The config file for TCRDock
+            It's should be a toml file with the keys listed in `envs`, including
+            `organism`, `mhc_class`, `mhc`, `peptide`, `va`, `ja`, `vb`, `jb`,
+            `cdr3a`, and `cdr3b`.
+            The values will overwrite the values in `envs`.
+
+    Output:
+        outdir: The output directory containing the results
+
+    Envs:
+        organism: The organism of the TCR, peptide and MHC
+        mhc_class (type=int): The MHC class, either `1` or `2`
+        mhc: The MHC allele, e.g., `A*02:01`
+        peptide: The peptide sequence
+        va: The V alpha gene
+        ja: The J alpha gene
+        vb: The V beta gene
+        jb: The J beta gene
+        cdr3a: The CDR3 alpha sequence
+        cdr3b: The CDR3 beta sequence
+        python: The path of python with dependencies for `tcrdock` installed.
+            If not provided, `TCRDock.lang` will be used (the same interpreter
+            used for the wrapper script).
+            It could also be a list to specify, for example, a python in a conda
+            environment (e.g., `["conda", "run", "-n", "myenv", "python"]`).
+        tmpdir: The temporary directory used to clone the `tcrdock` source code if
+            `envs.tcrdock` is not provided.
+        tcrdock: The path to the `tcrdock` source code repo.
+            You need to clone the source code from the github repository.
+            <https://github.com/phbradley/TCRdock> at
+            revision c5a7af42eeb0c2a4492a4d4fe803f1f9aafb6193 at main branch.
+            You also have to run `download_blast.py` after cloning to download the
+            blast database in the directory.
+            If not provided, we will clone the source code to the `envs.tmpdir`
+            directory and run the `download_blast.py` script.
+        model_name: The model name to use
+        model_file: The model file to use.
+            If provided as a relative path, it should be relative to the
+            `<envs.data_dir>/params/`, otherwise, it should be the full path.
+        data_dir: The data directory that contains the model files.
+            The model files should be in the `params` subdirectory.
+    """
+    input = "configfile:file"
+    output = "outdir:dir:{{in.configfile | stem}}.tcrdock"
+    lang = config.lang.python
+    envs = {
+        "tcrdock": None,
+        "organism": "human",
+        "mhc_class": 1,
+        "mhc": "A*02:01",
+        "peptide": None,
+        "va": None,
+        "ja": None,
+        "vb": None,
+        "jb": None,
+        "cdr3a": None,
+        "cdr3b": None,
+        "python": None,
+        "model_name": "model_2_ptm_ft4",
+        "model_file": "tcrpmhc_run4_af_mhc_params_891.pkl",
+        "data_dir": None,
+    }
+    script = "file://../scripts/tcr/TCRDock.py"

{biopipen-0.27.1 → biopipen-0.27.3}/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-0.27.1 → biopipen-0.27.3}/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-0.27.1 → biopipen-0.27.3}/biopipen/scripts/scrna/RadarPlots.R

@@ -74,10 +74,10 @@ expand_each <- function(name,  case) {
         }
     } else {
         if (is.null(case$subset)) {
-            eachs <- srtobj@meta.data %>%
+            eachs <- meta %>%
                 pull(case$each) %>% unique() %>% na.omit() %>% as.vector()
         } else {
-            eachs <- srtobj@meta.data %>% filter(!!parse_expr(case$subset)) %>%
+            eachs <- meta %>% filter(!!parse_expr(case$subset)) %>%
                 pull(case$each) %>% unique() %>% na.omit() %>% as.vector()
         }
         for (each in eachs) {

{biopipen-0.27.1 → biopipen-0.27.3}/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-0.27.1 → biopipen-0.27.3}/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-0.27.3/biopipen/scripts/snp/PlinkSimulation.py

@@ -0,0 +1,124 @@
+from pathlib import Path
+from multiprocessing import Pool
+from slugify import slugify
+from simpleconf import Config
+from biopipen.utils.misc import logger, run_command, dict_to_cli_args
+
+configfile = {{in.configfile | repr}}  # pyright: ignore # noqa: E999
+outdir = {{out.outdir | repr}}  # pyright: ignore
+gtmatfile = {{out.gtmat | repr}}  # pyright: ignore
+config = Config.load(configfile)
+
+default_nsnps = {{envs.nsnps | repr}}  # pyright: ignore
+default_ncases = {{envs.ncases | repr}}  # pyright: ignore
+default_nctrls = {{envs.nctrls | repr}}  # pyright: ignore
+default_plink = {{envs.plink | repr}}  # pyright: ignore
+default_seed = {{envs.seed | repr}}  # pyright: ignore
+default_label = {{envs.label | repr}}  # pyright: ignore
+default_prevalence = {{envs.prevalence | repr}}  # pyright: ignore
+default_minfreq = {{envs.minfreq | repr}}  # pyright: ignore
+default_maxfreq = {{envs.maxfreq | repr}}  # pyright: ignore
+default_hetodds = {{envs.hetodds | repr}}  # pyright: ignore
+default_homodds = {{envs.homodds | repr}}  # pyright: ignore
+default_missing = {{envs.missing | repr}}  # pyright: ignore
+default_args = {{envs.args | repr}}  # pyright: ignore
+default_transpose_gtmat = {{envs.transpose_gtmat | repr}}  # pyright: ignore
+default_sample_prefix = {{envs.sample_prefix | repr}}  # pyright: ignore
+
+defaults = {
+    "nsnps": default_nsnps,
+    "ncases": default_ncases,
+    "nctrls": default_nctrls,
+    "plink": default_plink,
+    "seed": default_seed,
+    "label": default_label,
+    "prevalence": default_prevalence,
+    "minfreq": default_minfreq,
+    "maxfreq": default_maxfreq,
+    "hetodds": default_hetodds,
+    "homodds": default_homodds,
+    "missing": default_missing,
+    # "args": default_args,
+    "transpose_gtmat": default_transpose_gtmat,
+    "sample_prefix": default_sample_prefix,
+}
+
+def do_one_simulation(confitems):
+    args = default_args.copy()
+    args.update(confitems.pop("args", {}))
+    confs = defaults.copy()
+    confs.update(confitems)
+    transpose_gtmat = confs.pop("transpose_gtmat")
+    sample_prefix = confs.pop("sample_prefix")
+
+
+    logger.debug("  Generating parameters file")
+    params_file = Path(outdir) / "params.txt"
+    params_file.write_text(
+        f"{confs['nsnps']}\t{confs['label']}\t{confs['minfreq']}\t"
+        f"{confs['maxfreq']}\t{confs['hetodds']}\t{confs['homodds']}\n"
+    )
+
+    if confs.get('seed') is not None:
+        args["seed"] = confs['seed']
+
+    args["simulate"] = params_file
+    args["out"] = Path(outdir) / "sim_snps"
+    args["simulate-ncases"] = confs['ncases']
+    args["simulate-ncontrols"] = confs['nctrls']
+    args["simulate-prevalence"] = confs['prevalence']
+    args["simulate-missing"] = confs['missing']
+
+    cmd = [confs['plink']] + dict_to_cli_args(args)
+
+    logger.debug("  Running PLINK simulation ...")
+    run_command(cmd, fg=True)
+
+    # Transpose the genotype matrix
+    # CHR	SNP	(C)M	POS	COUNTED	ALT	per0_per0	per1_per1	per2_per2
+    # 1	SNP_0	0	1	D	d	1	0	1
+    # 1	SNP_1	0	2	d	D	0	1	0
+    # 1	SNP_2	0	3	d	D	0	0	0
+    # 1	SNP_3	0	4	d	D	0	0	0
+    # 1	SNP_4	0	5	D	d	1	2	1
+    cmd = [
+        confs['plink'],
+        "--recode",
+        "A" if transpose_gtmat else "A-transpose",
+        "tab",
+        "--bfile",
+        args["out"],
+        "--out",
+        gtmatfile + ".plink.recoded",
+    ]
+    logger.debug("- Recoding into genotype matrix ...")
+    run_command(cmd, fg=True)
+
+    logger.debug("  Saving genotype matrix ...")
+    ## transpose_gtmat = False
+    # SNP_COUNTED	per0_per0	per1_per1	per2_per2
+    # SNP_0_D	1	0	1
+    # SNP_1_d	0	1	0
+    # SNP_2_d	0	0	0
+    # SNP_3_d	0	0	0
+    # SNP_4_D	1	2	1
+    ## transpose_gtmat = True
+    # FID_IID SNP_0_D SNP_1_D SNP_2_D
+    # per0_per0 0 1 1
+    # per1_per1 0 2 0
+    # per2_per2 0 0 0
+    # per3_per3 1 1 0
+    # per4_per4 0 0 0
+    if transpose_gtmat:
+        cmd = f"cut -f1,2,7- {gtmatfile}.plink.recoded.raw | sed 's/\\t/_/'"
+    else:
+        cmd = f"cut -f2,5,7- {gtmatfile}.plink.recoded.traw | sed 's/\\t/_/'"
+
+    if sample_prefix:
+        cmd = f"{cmd} | sed 's/per[0-9]\\+_per/{sample_prefix}/g'"
+
+    cmd = f"{cmd} > {gtmatfile}"
+    run_command(cmd, fg=True)
+
+
+do_one_simulation(config)

{biopipen-0.27.1 → biopipen-0.27.3}/biopipen/scripts/stats/DiffCoexpr.R

@@ -42,21 +42,21 @@ diffcoex_score <- function(group) {
 
     gvals <- unique(gdata[, group, drop = TRUE])
     if (length(gvals) < 2) {
-        log_warn("  Less than 2 groups in the input. Skipping ...")
+        log_debug("  Less than 2 groups in the input. Skipping ...")
         return(NULL)
     }
     rs <- lapply(gvals, function(gval) {
         samples <- rownames(gdata[gdata[[group]] == gval, , drop = FALSE])
         expr <- indata[samples, , drop = FALSE]
         if (length(samples) < 3) {
-            log_warn("  Less than 3 samples in one of the groups. Skipping ...")
+            log_debug("  Less than 3 samples in one of the groups. Skipping ...")
             return(NULL)
         }
         cor.pairs(as.matrix(expr), cor.method = method)
     })
     rs[sapply(rs, is.null)] <- NULL
     if (length(rs) < 2) {
-        log_warn("  Less than 2 groups with at least 3 samples. Skipping ...")
+        log_debug("  Less than 2 groups with at least 3 samples. Skipping ...")
         return(NULL)
     }
     N <- length(rs)