biopipen 0.33.0__py3-none-any.whl → 0.34.0__py3-none-any.whl

biopipen/reports/utils/gsea.liq

@@ -1,110 +0,0 @@
-{% from "utils/misc.liq" import table_of_images -%}
-
-{%- macro fgsea_report_script() -%}
-import { Image, DataTable } from "$libs";
-{%- endmacro -%}
-
-{%- macro fgsea_report(fgsea_dir, h, envs, nrows=100) -%}
-{%- addfilter splitgenes -%}
-def splitgenes(data):
-    for dat in data:
-        dat["leadingEdge"] = dat["leadingEdge"].replace(",", " ")
-    return json_dumps(data)
-{%- endaddfilter -%}
-
-<h{{h}}>Enrichment table</h{{h}}>
-<Image src={{ fgsea_dir | joinpaths: "gsea_table.png" | quote }} />
-
-{% set data = fgsea_dir | joinpaths: "fgsea.txt" | datatable: sep="\t", nrows=nrows | json_loads %}
-
-<h{{h}}>Enrichment pathways</h{{h}}>
-<DataTable src={{ fgsea_dir | joinpaths: "fgsea.txt" | quote }}
-    data={ {{ data | splitgenes: }} }
-    pageSize={10} />
-
-<h{{h}}>Enrichment plot of pathways</h{{h}}>
-{%- python -%}
-import os
-def fgsea_plots(pathways, fgsea_dir):
-    out = []
-    for pathway in pathways:
-        pathway = pathway.replace("/", "-")
-        pwfig = joinpaths(fgsea_dir, f"fgsea_{pathway}.png")
-        if os.path.exists(pwfig):
-            out.append(pwfig)
-    return out
-{%- endpython -%}
-{{ table_of_images(
-    fgsea_plots(liquid_map(data, "pathway"), fgsea_dir),
-    liquid_map(data, "pathway"),
-    table_width=75
-) }}
-
-{%- endmacro -%}
-
-
-{%- macro gsea_report(gsea_dir, h, envs, nrows=100) -%}
-<h{{h}}>Global view</h{{h}}>
-
-<embed src={{gsea_dir | joinpaths: "*.global.plots.pdf" | glob | first | quote}}
-    width="100%"
-    height="1000"
-    type="application/pdf" />
-
-<h{{h}}>Summary</h{{h}}>
-{% for sumfile in gsea_dir | joinpaths: "*.SUMMARY.RESULTS.REPORT.*.txt" | glob %}
-{%   set klass = stem(sumfile).split(".")[-1] %}
-<h{{h+1}}>{{klass}}</h{{h+1}}>
-<DataTable data={ {{sumfile | datatable: sep="\t", nrows=nrows}} } />
-{% endfor %}
-
-<h{{h}}>Enrichment details</h{{h}}>
-{% set cutoff = envs.get("fdr.q.val.threshold", envs.get("fdr_q_val_threshold", 0.25)) %}
-{% for sumfile in gsea_dir | joinpaths: "*.SUMMARY.RESULTS.REPORT.*.txt" | glob %}
-{%   set klass = stem(sumfile).split(".")[-1] %}
-<h{{h+1}}>{{klass}}</h{{h+1}}>
-{%   set sumdata = sumfile | datatable: sep="\t" | json_loads %}
-{%   set has_signif = [] %}
-{%   for row in sumdata %}
-{%      if row["FDR_q_val"] < cutoff %}
-{%          set _ = has_signif.append(1) %}
-<embed src={{gsea_dir | joinpaths: "*." + row["GS"] + ".plot." + klass + ".*.pdf" | glob | first | quote}}
-    width="100%"
-    height="700"
-    type="application/pdf" />
-{%      endif %}
-{%   endfor %}
-{%   if len(has_signif) == 0 %}
-<Tile>No significantly (FDR_q_val &lt; {{cutoff}}) enriched pathways found.</Tile>
-{%   endif %}
-{% endfor %}
-
-{%- endmacro -%}
-
-
-{%- macro enrichr_report_script() -%}
-import { Image, DataTable } from "$libs";
-import { Tabs, Tab, TabContent, InlineNotification } from "$ccs";
-{%- endmacro -%}
-
-{%- macro enrichr_report(enrichr_dir) -%}
-<Tabs>
-    {% for enrtxt in enrichr_dir | glob: "Enrichr-*.txt"  %}
-        {% set db = enrtxt | stem | replace: "Enrichr-", "" %}
-        <Tab label="{{db}}" title="{{db}}" />
-    {% endfor %}
-    <div slot="content">
-        {% for enrtxt in enrichr_dir | glob: "Enrichr-*.txt" %}
-            {% set db = enrtxt | stem | replace: "Enrichr-", "" %}
-            <TabContent>
-                <Image src={{enrichr_dir | joinpaths: "Enrichr-" + db + ".png" | quote}} />
-                <DataTable
-                    src={{ enrtxt | quote }}
-                    data={ {{ enrtxt | datatable: sep="\t", nrows=100 }} }
-                    />
-            </TabContent>
-        {% endfor %}
-    </div>
-</Tabs>
-{%- endmacro -%}
-

biopipen/scripts/scrna/CellTypeAnnotation-common.R

@@ -1,10 +0,0 @@
-merge_clusters_with_same_labels <- function(sobj, newcol) {
-    if (is.null(newcol)) {
-        sobj@meta.data$seurat_clusters <- sub("\\.\\d+$", "", sobj@meta.data$seurat_clusters)
-        Idents(sobj) <- "seurat_clusters"
-    } else {
-        sobj@meta.data[[newcol]] <- sub("\\.\\d+$", "", sobj@meta.data[[newcol]])
-    }
-
-    sobj
-}

biopipen/scripts/scrna/SeuratClustering-common.R

@@ -1,213 +0,0 @@
-
-expand_dims <- function(args, name = "dims") {
-    # Expand dims from 30 to 1:30
-    if (is.numeric(args[[name]]) && length(args[[name]] == 1)) {
-        args[[name]] <- 1:args[[name]]
-    }
-    args
-}
-
-expand_resolution <- function(resolution) {
-    expanded_res <- c()
-    for (res in resolution) {
-        if (is.numeric(res)) {
-            expanded_res <- c(expanded_res, res)
-        } else {
-            # is.character
-            parts <- trimws(unlist(strsplit(res, ",")))
-            for (part in parts) {
-                if (grepl(":", part)) {
-                    ps <- trimws(unlist(strsplit(part, ":")))
-                    if (length(ps) == 2) { ps <- c(ps, 0.1) }
-                    if (length(ps) != 3) {
-                        stop("Invalid resolution format: {part}. Expected 2 or 3 parts separated by ':' for a range.")
-                    }
-                    ps <- as.numeric(ps)
-                    expanded_res <- c(expanded_res, seq(ps[1], ps[2], by = ps[3]))
-                } else {
-                    expanded_res <- c(expanded_res, as.numeric(part))
-                }
-            }
-        }
-    }
-    # keep the last resolution at last
-    rev(unique(rev(round(expanded_res, 2))))
-}
-
-# recode clusters from 0, 1, 2, ... to c1, c2, c3, ...
-recode_clusters <- function(clusters) {
-    recode <- function(x) paste0("c", as.integer(as.character(x)) + 1)
-    clusters <- factor(recode(clusters), levels = recode(levels(clusters)))
-    clusters
-}
-
-run_transformation <- function(sobj) {
-    if (length(envs$ScaleData) == 0 && length(envs$SCTransform) == 0) {
-        log_warn("Skipping ScaleData/SCTransform (neither specified) ...")
-        return(sobj)
-    }
-    if (length(envs$ScaleData) > 0 && length(envs$SCTransform) > 0) {
-        stop("Both envs.ScaleData and envs.SCTransform are specified. Please choose either.")
-    }
-    if (length(envs$ScaleData) > 0) {
-        if (DefaultAssay(sobj) == "SCT") {
-            stop("SCT assay detected, but envs.ScaleData is specified. Use envs.SCTransform instead.")
-        }
-        cached <- get_cached(envs$ScaleData, "ScaleData", cache_dir)
-        if (is.null(cached$data)) {
-            log_info("Running ScaleData ...")
-            sobj <- do_call(ScaleData, c(list(object = sobj), envs$ScaleData))
-            cached$data <- list(assay = sobj@assays$RNA, commands = sobj@commands)
-            save_to_cache(cached, "ScaleData", cache_dir)
-        } else {
-            log_info("Loading cached ScaleData ...")
-            sobj@assays$RNA <- cached$data$assay
-            sobj@commands <- cached$data$commands
-            DefaultAssay(sobj) <- "RNA"
-        }
-    } else if (length(envs$SCTransform) > 0) {
-        if (DefaultAssay(sobj) != "SCT") {
-            stop("SCT assay not detected, but envs.SCTransform is specified. Use envs.ScaleData instead.")
-        }
-        cached <- get_cached(envs$SCTransform, "SCTransform", cache_dir)
-        asssay <- envs$SCTransform$new.assay.name %||% "SCT"
-        if (is.null(cached$data)) {
-            log_info("Running SCTransform ...")
-            sobj <- do_call(SCTransform, c(list(object = sobj), envs$SCTransform))
-            cached$data <- list(assay = sobj@assays$SCT, commands = sobj@commands)
-            save_to_cache(cached, "SCTransform", cache_dir)
-        } else {
-            log_info("Loading cached SCTransform ...")
-            sobj@assays[[assay]] <- cached$data$assay
-            sobj@commands <- cached$data$commands
-            DefaultAssay(sobj) <- assay
-        }
-    }
-    sobj
-}
-
-run_umap <- function(sobj) {
-    cached <- get_cached(
-        list(sobj = sobj, RunUMAP = envs$RunUMAP),
-        "RunUMAP",
-        cache_dir
-    )
-    reduc_name <- envs$RunUMAP$reduction.name %||% "umap"
-    if (is.null(cached$data)) {
-        log_info("Running RunUMAP ...")
-        umap_args <- list_setdefault(
-            envs$RunUMAP,
-            object = sobj,
-            dims = 1:30,
-            reduction = sobj@misc$integrated_new_reduction %||% "pca"
-        )
-        ncells <- ncol(sobj)
-        umap_args$dims <- 1:min(max(umap_args$dims), ncells - 1)
-        umap_method <- envs$RunUMAP$umap.method %||% "uwot"
-        if (umap_method == "uwot" && is.null(envs$RunUMAP$n.neighbors)) {
-            # https://github.com/satijalab/seurat/issues/4312
-            umap_args$n.neighbors <- min(ncells - 1, 30)
-        }
-        sobj <- do_call(RunUMAP, umap_args)
-        cached$data <- list(reduc = sobj@reductions[[reduc_name]], commands = sobj@commands)
-        save_to_cache(cached, "RunUMAP", cache_dir)
-    } else {
-        log_info("Loading cached RunUMAP ...")
-        sobj@reductions[[reduc_name]] <- cached$data$reduc
-        sobj@commands <- cached$data$commands
-    }
-
-    sobj
-}
-
-run_findneighbors <- function(sobj) {
-    cached <- get_cached(
-        list(sobj = sobj, FindNeighbors = envs$FindNeighbors),
-        "FindNeighbors",
-        cache_dir
-    )
-    if (is.null(cached$data)) {
-        log_info("Running FindNeighbors ...")
-        envs$FindNeighbors$object <- sobj
-        envs$FindNeighbors$reduction <- sobj@misc$integrated_new_reduction %||% "pca"
-        sobj <- do_call(FindNeighbors, envs$FindNeighbors)
-        cached$data <- list(graphs = sobj@graphs, commands = sobj@commands)
-        save_to_cache(cached, "FindNeighbors", cache_dir)
-    } else {
-        log_info("Loading cached FindNeighbors ...")
-        sobj@graphs <- cached$data$graphs
-        sobj@commands <- cached$data$commands
-    }
-
-    sobj
-}
-
-run_findclusters <- function(sobj) {
-    cached <- get_cached(
-        list(sobj = sobj, FindClusters = envs$FindClusters),
-        "FindClusters",
-        cache_dir
-    )
-    if (is.null(cached$data)) {
-        findclusters_args <- envs$FindClusters
-        findclusters_args$random.seed <- findclusters_args$random.seed %||% 8525
-        resolution <- findclusters_args$resolution <- expand_resolution(findclusters_args$resolution %||% 0.8)
-        log_info("Running FindClusters at resolution: {paste(resolution, collapse=',')} ...")
-
-        findclusters_args$object <- sobj
-        findclusters_args$cluster.name <- paste0("seurat_clusters.", resolution)
-        sobj <- do_call(FindClusters, findclusters_args)
-
-        for (clname in findclusters_args$cluster.name) {
-            sobj@meta.data[[clname]] <- recode_clusters(sobj@meta.data[[clname]])
-        }
-        sobj@meta.data$seurat_clusters <- recode_clusters(sobj@meta.data$seurat_clusters)
-        Idents(sobj) <- "seurat_clusters"
-
-        ident_table <- table(Idents(sobj))
-        log_info("- Found {length(ident_table)} clusters at resolution {resolution[length(resolution)]}")
-        print(ident_table)
-        cat("\n")
-
-        cached$data <- list(
-            clusters = sobj@meta.data[, c(findclusters_args$cluster.name, "seurat_clusters"), drop = FALSE],
-            commands = sobj@commands
-        )
-        save_to_cache(cached, "FindClusters", cache_dir)
-    } else {
-        log_info("Loading cached FindClusters ...")
-
-        sobj <- AddMetaData(sobj, metadata = cached$data$clusters)
-        Idents(sobj) <- "seurat_clusters"
-        sobj@commands <- cached$data$commands
-    }
-
-    sobj
-}
-
-run_prepsctfindmarkers <- function(sobj) {
-    if (DefaultAssay(sobj) == "SCT") {
-        cached <- get_cached(list(sobj = sobj), "PrepSCTFindMarkers", cache_dir)
-        if (is.null(cached$data)) {
-            # https://github.com/satijalab/seurat/issues/6968
-            log_info("Running PrepSCTFindMarkers ...")
-            sobj <- PrepSCTFindMarkers(sobj)
-            # compose a new SeuratCommand to record it to sobj@commands
-            scommand <- sobj@commands$FindClusters
-            scommand@name <- "PrepSCTFindMarkers"
-            scommand@time.stamp <- Sys.time()
-            scommand@assay.used <- "SCT"
-            scommand@call.string <- "PrepSCTFindMarkers(object = sobj)"
-            scommand@params <- list()
-            sobj@commands$PrepSCTFindMarkers <- scommand
-
-            cached$data <- sobj
-            save_to_cache(cached, "PrepSCTFindMarkers", cache_dir)
-        } else {
-            log_info("Loading cached PrepSCTFindMarkers ...")
-            sobj <- cached$data
-        }
-    }
-
-    sobj
-}

biopipen/scripts/scrna_metabolic_landscape/MetabolicFeaturesIntraSubset.R

@@ -1,193 +0,0 @@
-{{ biopipen_dir | joinpaths: "utils", "misc.R" | source_r }}
-{{ biopipen_dir | joinpaths: "utils", "gsea.R" | source_r }}
-
-library(parallel)
-library(scater)
-library(Seurat)
-
-sobjfile <- {{ in.sobjfile | r }}
-outdir <- {{ out.outdir | r }}
-joboutdir <- {{ job.outdir | r }}
-gmtfile <- {{ envs.gmtfile | r }}
-ncores <- {{ envs.ncores | r }}
-fgsea <- {{ envs.fgsea | r }}
-top <- {{ envs.top | r }}
-prerank_method <- {{ envs.prerank_method | r }}
-grouping <- {{ envs.grouping | r }}
-grouping_prefix <- {{ envs.grouping_prefix | r }}
-subsetting_cols <- {{ envs.subsetting | r }}
-subsetting_prefix <- {{ envs.subsetting_prefix | r }}
-subsetting_comparison <- {{ envs.subsetting_comparison | r }}
-
-if (!is.null(grouping_prefix) && nchar(grouping_prefix) > 0) {
-    grouping_prefix = paste0(grouping_prefix, "_")
-}
-
-if (!is.null(subsetting_prefix) && nchar(subsetting_prefix) > 0) {
-    subsetting_prefix = paste0(subsetting_prefix, "_")
-}
-
-set.seed(8525)
-
-## gmt_pathways is copied from fgsea package.
-gmt_pathways <- function(gmt_file) {
-    pathway_lines <- strsplit(readLines(gmt_file), "\t")
-    pathways <- lapply(pathway_lines, tail, -2)
-    names(pathways) <- sapply(pathway_lines, head, 1)
-    pathways
-}
-
-gmtfile <- localizeGmtfile(gmtfile)
-pathways <- gmt_pathways(gmtfile)
-metabolics <- unique(as.vector(unname(unlist(pathways))))
-sobj <- readRDS(sobjfile)
-
-do_one_comparison <- function(
-    obj,
-    compname,
-    genes,
-    case,
-    control,
-    groupdir,
-    subset_col,
-    subset_prefix,
-    groupname
-) {
-    log_info(paste("  Design: {compname} ({case}, {control})"))
-    case_code = paste0("subset(obj, subset = ", subset_col, " == '", case, "')")
-    case_obj = tryCatch({
-        eval(parse(text = case_code))
-    }, error = function(e) {
-        NULL
-    })
-    if (is.null(case_obj)) {
-        log_warn("          Skip (not enough cells in case)")
-        return (NULL)
-    }
-    control_code = paste0("subset(obj, subset = ", subset_col, " == '", control, "')")
-    control_obj = tryCatch({
-        eval(parse(text = control_code))
-    }, error = function(e) {
-        NULL
-    })
-    if (is.null(control_obj)) {
-        log_warn("          Skip (not enough cells in control)")
-        add_report(
-            list(kind = "error", content = "Not enough cells in control"),
-            h1 = groupname,
-            h2 = compname
-        )
-        return (NULL)
-    }
-    exprs_case = GetAssayData(case_obj)[genes, , drop = FALSE]
-    exprs_control = GetAssayData(control_obj)[genes, , drop = FALSE]
-
-    odir = file.path(groupdir, paste0(subset_prefix, compname))
-    dir.create(odir, showWarnings = FALSE)
-    if (ncol(exprs_case) < 5 || ncol(exprs_control) < 5) {
-        log_warn("  Skipped (not enough cells).")
-        wfile <- file.path(odir, "warning.txt")
-        write("Skipped (not enough cells)\n\n", file = wfile)
-        write(paste0("n_cells (Case):", ncol(exprs_case)), file = wfile, append = TRUE)
-        write(paste0("n_cells (Control):", ncol(exprs_control)), file = wfile, append = TRUE)
-
-        return(list(
-            list(kind = "error", content = "Not enough cells"),
-            h1 = groupname,
-            h2 = compname
-        ))
-    }
-    if (fgsea) {
-        ranks = prerank(
-            cbind(exprs_case, exprs_control),
-            case,
-            control,
-            c(rep(case, ncol(exprs_case)), rep(control, ncol(exprs_control))),
-            method = prerank_method
-        )
-
-        runFGSEA(
-            ranks,
-            gmtfile,
-            top = top,
-            outdir = odir,
-            envs = list(nproc = 1)
-        )
-
-        report = list(
-            list(kind = "fgsea", dir = odir),
-            h1 = groupname,
-            h2 = compname
-        )
-    } else {
-        runGSEA(
-            cbind(exprs_case, exprs_control),
-            c(rep(case, ncol(exprs_case)), rep(control, ncol(exprs_control))),
-            gmtfile,
-            odir
-        )
-
-        report = list()
-    }
-
-    report
-}
-
-do_one_group <- function(group) {
-    log_info("- Group: {group} ...")
-
-    genes = intersect(metabolics, rownames(sobj))
-    group_code = paste0(
-        "subset(sobj, subset = ", grouping, " == '", group, "')"
-    )
-    obj = eval(parse(text = group_code))
-    groupname = paste0(grouping_prefix, group)
-    groupdir = file.path(outdir, slugify(groupname))
-    dir.create(groupdir, showWarnings = FALSE)
-
-    report = list()
-    for (i in seq_along(subsetting_comparison)) {
-        sci = subsetting_comparison[[i]]
-        if (is.null(sci) || length(sci) == 0) {
-            next
-        }
-        rs = lapply(
-            names(sci),
-            function(compname) {
-                do_one_comparison(
-                    obj,
-                    compname,
-                    genes,
-                    sci[[compname]][1],
-                    sci[[compname]][2],
-                    groupdir,
-                    subsetting_cols[i],
-                    subsetting_prefix[i],
-                    groupname
-                )
-            }
-        )
-        if (length(rs) > 0) {
-            report = c(report, rs)
-        }
-    }
-    report
-}
-
-groups = sort(as.character(unique(sobj@meta.data[[grouping]])))
-if (ncores == 1) {
-    x = lapply(groups, do_one_group)
-} else {
-    x = mclapply(groups, do_one_group, mc.cores = ncores)
-    if (any(unlist(lapply(x, class)) == "try-error")) {
-        stop("mclapply error")
-    }
-}
-report = unlist(x, recursive = FALSE)
-for (r in report) {
-    if (!is.null(r)) {
-        do.call(add_report, r)
-    }
-}
-
-save_report(joboutdir)

biopipen/utils/caching.R

@@ -1,44 +0,0 @@
-library(digest)
-
-#' Get signatures and cached data
-#'
-#' @param x An object to infer signature from
-#' @param kind A string indicating the kind of the object
-#'   Used as part of the filename of the cached file
-#' @param cache_dir A string indicating the directory to store cached files
-#'
-#' @return A list containing the signature, digested signature and cached data
-get_cached <- function(x, kind, cache_dir) {
-    if (is.null(cache_dir) || isFALSE(cache_dir)) {
-        return(list(sig = NULL, dig = NULL, data = NULL))
-    }
-    # Get signature of an object
-    sig <- capture.output(str(x))
-    dig <- digest::digest(sig, algo = "md5")
-    dig <- substr(dig, 1, 8)
-    cached_file <- file.path(cache_dir, paste0(dig, ".", kind, ".RDS"))
-    if (!file.exists(cached_file)) {
-        return(list(sig = sig, dig = dig, data = NULL))
-    }
-
-    list(sig = sig, dig = dig, data = readRDS(cached_file))
-}
-
-#' Save an object to cache
-#'
-#' @param to_cache An list to cache,
-#'   including the signature, digested signature and data
-#' @param kind A string indicating the kind of the object
-#'   Used as part of the filename of the cached file
-#' @param cache_dir A string indicating the directory to store cached files
-save_to_cache <- function(to_cache, kind, cache_dir) {
-    if (is.null(cache_dir) || isFALSE(cache_dir)) { return() }
-    dig <- to_cache$dig
-    sig <- to_cache$sig
-    data <- to_cache$data
-    # Save an object to cache
-    sig_file <- file.path(cache_dir, paste0(dig, ".", kind , ".signature.txt"))
-    writeLines(c(as.character(Sys.time()), "", sig), sig_file)
-    cached_file <- file.path(cache_dir, paste0(dig, ".", kind, ".RDS"))
-    saveRDS(data, cached_file)
-}

biopipen/utils/gene.R

@@ -1,95 +0,0 @@
-suppressPackageStartupMessages({
-    library(rlang)
-    library(dplyr)
-    library(mygene)
-})
-
-
-#@' Convert gene names between different formats
-#@'
-#@' @param genes A character/integer vector of gene names/ids
-#@' @param species A character vector of species names
-#@' @param infmt A character vector of input gene name formats
-#@'   See the available scopes at
-#@'   https://docs.mygene.info/en/latest/doc/data.html#available-fields
-#@'   You can use ensg as a shortcut for ensembl.gene
-#@' @param outfmt A character vector of output gene name formats
-#@' @param dup How to deal with duplicate gene names found.
-#@'   "first": keep the first one (default), sorted by score descendingly
-#@'   "last": keep the last one, sorted by score descendingly
-#@'   "all": keep all of them, each will be a separate row
-#@'   "<X>": combine them into a single string, separated by X
-#@' @param notfound How to deal with gene names that are not found
-#@'   "error": stop with an error message
-#@'   "use-query": use the query gene name as the converted gene name
-#@'   "skip": skip the gene names that are not found
-#@'   "ignore": Same as "skip"
-#@'   "na": use NA as the converted gene name (default)
-#@' @param suppress_messages Whether to suppress the warning messages
-#@' @return A tibble with the query gene names and the converted gene names
-#@'   When a gene name is not found, the converted name will be NA
-#@'   When duplicate gene names are found, the one with the highest score will be kept
-#@' @export
-gene_name_conversion <- function(
-    genes,
-    infmt,
-    outfmt,
-    dup = "first",
-    species = "human",
-    notfound = "na",
-    suppress_messages = FALSE
-) {
-    notfound <- arg_match(notfound, c("error", "use-query", "skip", "ignore", "na"))
-
-    if (infmt %in% c("ensg", "ensmusg")) { infmt = "ensembl.gene" }
-    if (outfmt %in% c("ensg", "ensmusg")) { outfmt = "ensembl.gene" }
-
-    orig_genes <- genes
-    if (infmt == "ensembl.gene") {
-        # Remove version numbers from ensembl gene ids
-        genes <- gsub("\\..*", "", genes)
-    }
-    query_df <- tibble(query = genes, orig = orig_genes)
-
-    if (suppress_messages) {
-        capture.output(suppressWarnings(suppressMessages({
-            out <- queryMany(genes, scopes=infmt, fields=outfmt, species=species) %>%
-                as_tibble()
-        })))
-    } else {
-        out <- queryMany(genes, scopes=infmt, fields=outfmt, species=species) %>%
-            as_tibble()
-    }
-
-    if (nrow(out) == 0) {
-        return(tibble(query = orig_genes, converted = NA_character_))
-    }
-
-    if (dup == "first") {
-        out = out %>% group_by(query) %>% arrange(desc(X_score)) %>%
-            slice_head(n=1) %>% ungroup() %>% dplyr::select(all_of(c("query", outfmt)))
-    } else if (dup == "last") {
-        out = out %>% group_by(query) %>% arrange(X_score) %>%
-            slice_head(n=1) %>% ungroup() %>% dplyr::select(all_of(c("query", outfmt)))
-    } else if (dup != "all") {
-        out = out %>% group_by(query) %>% arrange(desc(X_score)) %>%
-            summarise(!!sym(outfmt) := paste(unique(!!sym(outfmt)), collapse=dup))
-    }
-    out <- query_df %>%
-        left_join(out, by="query") %>%
-        dplyr::select(-"query") %>%
-        dplyr::select(query = orig, everything())
-
-    if (notfound == "error") {
-        if (any(is.na(out[[outfmt]]))) {
-            nagenes = out %>% filter(is.na(.[[outfmt]])) %>% pull("query")
-            stop(paste("Query genes not found:", paste(nagenes, collapse=",")))
-        }
-    } else if (notfound == "use-query") {
-        out = out %>% mutate(!!sym(outfmt) := coalesce(!!sym(outfmt), query))
-    } else if (notfound == "skip" || notfound == "ignore") {
-        out = out %>% filter(!is.na(!!sym(outfmt)))
-    }
-
-    return(out)
-}