biopipen 0.27.0__tar.gz → 0.27.2__tar.gz

{biopipen-0.27.0 → biopipen-0.27.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: biopipen
-Version: 0.27.0
+Version: 0.27.2
 Summary: Bioinformatics processes/pipelines that can be run from `pipen run`
 License: MIT
 Author: pwwang

biopipen-0.27.2/biopipen/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.27.2"

{biopipen-0.27.0 → biopipen-0.27.2}/biopipen/ns/scrna.py

@@ -1761,14 +1761,12 @@ class SeuratMap2Ref(Proc):
             Used in `future::plan(strategy = "multicore", workers = <ncores>)`
             to parallelize some Seurat procedures.
             See also: <https://satijalab.org/seurat/articles/future_vignette.html>
-        use: A column name (e.g. `predicted.celltype.l1`, `predicted.celltype.l2`)
-            to use as the final cell types in downstream analysis, which will be stored
-            in the `name` column and set as the `Idents` of the seurat object.
-            Or an expression to pass to `mutate()` to create a new column.
-            When not specified, `predicted.<key>` will be used if
-            `envs.MapQuery.refdata` has only one key, otherwise an error will be raised.
-        name: The name of the final cell type column.
-            This will be set as the `Idents` of the seurat object.
+        use: A column name of metadata from the reference
+            (e.g. `celltype.l1`, `celltype.l2`) to transfer to the query as the
+            cell types (ident) for downstream analysis. This field is required.
+            If you want to transfer multiple columns, you can use
+            `envs.MapQuery.refdata`.
+        ident: The name of the ident for query transferred from `envs.use` of the reference.
         ref: The reference seurat object file.
             Either an RDS file or a h5seurat file that can be loaded by
             `Seurat::LoadH5Seurat()`.
@@ -1790,8 +1788,8 @@ class SeuratMap2Ref(Proc):
                 Note that the hyphen (`-`) will be transformed into `.` for the keys.
         MapQuery (ns): Arguments for [`MapQuery()`](https://satijalab.org/seurat/reference/mapquery)
             - reference-reduction: Name of reduction to use from the reference for neighbor finding
-            - reduction-model: `DimReduc` object that contains the umap model
-            - refdata (type=json): Data to transfer
+            - reduction-model: `DimReduc` object that contains the umap model.
+            - refdata (type=json): Extra data to transfer from the reference to the query.
             - <more>: See <https://satijalab.org/seurat/reference/mapquery>.
                 Note that the hyphen (`-`) will be transformed into `.` for the keys.
         MappingScore (ns): Arguments for [`MappingScore()`](https://satijalab.org/seurat/reference/mappingscore)
@@ -1807,9 +1805,8 @@ class SeuratMap2Ref(Proc):
     lang = config.lang.rscript
     envs = {
         "ncores": config.misc.ncores,
-        # "use": "predicted.celltype.l2",
         "use": None,
-        "name": "seurat_clusters",
+        "ident": "seurat_clusters",
         "ref": None,
         "SCTransform": {
             "do-correct-umi": False,

{biopipen-0.27.0 → biopipen-0.27.2}/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-0.27.0 → biopipen-0.27.2}/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-0.27.0 → biopipen-0.27.2}/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-0.27.0 → biopipen-0.27.2}/biopipen/scripts/scrna/SeuratMap2Ref.R

@@ -10,7 +10,7 @@ set.seed(8525)
 sobjfile = {{in.sobjfile | r}}
 outfile = {{out.outfile | r}}
 use = {{envs.use | r}}
-name = {{envs.name | r}}
+ident = {{envs.ident | r}}
 ref = {{envs.ref | r}}
 ncores = {{envs.ncores | r}}
 sctransform_args = {{envs.SCTransform | r: todot="-"}}
@@ -18,23 +18,21 @@ findtransferanchors_args = {{envs.FindTransferAnchors | r: todot="-"}}
 mappingscore_args = {{envs.MappingScore | r: todot="-"}}
 mapquery_args = {{envs.MapQuery | r: todot="-"}}
 
-if (is.null(mapquery_args$refdata) || length(mapquery_args$refdata) == 0) {
-    stop("No refdata provided for MapQuery (envs.MapQuery.refdata)")
-}
-
 # See if we have a reference
 if (is.null(ref)) {
     stop("No reference provided (envs.ref)")
 }
 
 if (is.null(use)) {
-    if (length(mapquery_args$refdata) == 1) {
-        use = paste0("predicted.", names(mapquery_args$refdata)[1])
-    } else {
-        stop("No use provided (envs.use), don't know which column to use as cluster")
-    }
+    stop("No use provided (envs.use), don't know which column to transfer as cluster")
+}
+
+if (is.null(mapquery_args$refdata) || length(mapquery_args$refdata) == 0) {
+    mapquery_args$refdata = list()
 }
 
+mapquery_args$refdata[[use]] = use
+
 outdir = dirname(outfile)
 options(future.globals.maxSize = 80000 * 1024^2)
 plan(strategy = "multicore", workers = ncores)
@@ -49,7 +47,7 @@ plan(strategy = "multicore", workers = ncores)
 findtransferanchors_args = .expand_dims(findtransferanchors_args)
 
 # Load reference
-log_info("Loading reference")
+log_info("- Loading reference")
 if (endsWith(ref, ".rds") || endsWith(ref, ".RDS")) {
     reference = readRDS(ref)
 } else if (endsWith(ref, ".h5ad") || endsWith(ref, ".H5AD")) {
@@ -59,30 +57,30 @@ if (endsWith(ref, ".rds") || endsWith(ref, ".RDS")) {
 }
 
 # Load Seurat object
-log_info("Loading Seurat object")
+log_info("- Loading Seurat object")
 sobj = readRDS(sobjfile)
 
 # Normalize data
-log_info("Normalizing data")
+log_info("- Normalizing data")
 sctransform_args$object = sobj
 sctransform_args$residual.features = rownames(x = reference)
 query = do_call(SCTransform, sctransform_args)
 
 # Find anchors between query and reference
-log_info("Finding anchors")
+log_info("- Finding anchors")
 findtransferanchors_args$reference = reference
 findtransferanchors_args$query = query
 anchors = do_call(FindTransferAnchors, findtransferanchors_args)
 
 # Map query to reference
-log_info("Mapping query to reference")
+log_info("- Mapping query to reference")
 mapquery_args$reference = reference
 mapquery_args$query = query
 mapquery_args$anchorset = anchors
 query = do_call(MapQuery, mapquery_args)
 
 # Calculating mapping score
-log_info("Calculating mapping score")
+log_info("- Calculating mapping score")
 mappingscore_args$anchors = anchors
 mappingscore = tryCatch({
     do_call(MappingScore, mappingscore_args)
@@ -98,7 +96,7 @@ mappingscore = tryCatch({
 })
 
 # Calculate mapping score and add to metadata
-log_info("Calculating mapping score")
+log_info("- Calculating mapping score")
 query = AddMetaData(
   object = query,
   metadata = mappingscore,
@@ -106,12 +104,14 @@ query = AddMetaData(
 )
 
 # Add the alias to the metadata for the clusters
-log_info("Adding name to metadata and set as ident")
-query@meta.data = query@meta.data %>% mutate(!!sym(name) := as.factor(!!parse_expr(use)))
-Idents(query) = name
+log_info("- Adding ident to metadata and set as ident")
+query@meta.data = query@meta.data %>% mutate(
+    !!sym(ident) := as.factor(!!parse_expr(paste0("predicted.", use)))
+)
+Idents(query) = ident
 
 # Save
-log_info("Saving")
+log_info("- Saving result ...")
 saveRDS(query, file = outfile)
 
 
@@ -120,36 +120,37 @@ saveRDS(query, file = outfile)
 # ############################
 
 # # Plot the UMAP
-log_info("Plotting")
-for (refname in names(mapquery_args$refdata)) {
-    if (refname == "predicted_ADT") {
+log_info("- Plotting for transferred data ...")
+ref.reduction = mapquery_args$reduction.model %||% "wnn.umap"
+for (qname in names(mapquery_args$refdata)) {
+    rname <- mapquery_args$refdata[[qname]]
+
+    if (grepl("Array", class(reference[[rname]])) && grepl("Array", class(query[[qname]]))) {
+        log_warn("  Skipping transferred array: {qname} -> {rname}")
         next
     }
-    reduction = if (is.null(mapquery_args$reduction.model)) "wnn.umap" else mapquery_args$reduction.model
-    print(paste0("Plotting UMAP for ", refname))
-    p = DimPlot(
+
+    log_info("  Plotting transferred data: {qname} -> {rname}")
+
+    ref_p <- DimPlot(
         object = reference,
-        reduction = reduction,
-        group.by = refname,
+        reduction = ref.reduction,
+        group.by = rname,
         label = TRUE,
         label.size = 3,
         repel = TRUE,
     ) + NoLegend()
 
-    png(file.path(outdir, paste0("Reference_UMAP_", refname, ".png")), width = 1000, height = 1000, res = 100)
-    print(p)
-    dev.off()
-
-    p = DimPlot(
+    query_p <- DimPlot(
         object = query,
         reduction = "ref.umap",
-        group.by = paste0("predicted.", refname),
+        group.by = paste0("predicted.", qname),
         label = TRUE,
         label.size = 3,
         repel = TRUE,
     ) + NoLegend()
 
-    png(file.path(outdir, paste0("Query_UMAP_", refname, ".png")), width = 1000, height = 1000, res = 100)
-    print(p)
+    png(file.path(outdir, paste0("UMAPs.png")), width = 1400, height = 700, res = 100)
+    print(ref_p | query_p)
     dev.off()
 }

biopipen-0.27.2/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-0.27.0 → biopipen-0.27.2}/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)

{biopipen-0.27.0 → biopipen-0.27.2}/biopipen/scripts/tcr/CloneResidency.R

@@ -26,6 +26,7 @@ section <- {{ envs.section | r }}
 mutaters <- {{ envs.mutaters | r }}
 subset <- {{ envs.subset | r }}
 prefix <- {{ envs.prefix | r }}
+upset_ymax <- {{ envs.upset_ymax | r }}
 upset_trans <- {{ envs.upset_trans | r }}
 cases <- {{ envs.cases | r }}
 
@@ -40,6 +41,7 @@ if (is.null(cases) || length(cases) == 0) {
             order = sample_order,
             subset = subset,
             section = section,
+            upset_ymax = upset_ymax,
             upset_trans = upset_trans
         )
     )
@@ -50,6 +52,7 @@ if (is.null(cases) || length(cases) == 0) {
         cases[[key]]$order <- cases[[key]]$order %||% sample_order
         cases[[key]]$section <- cases[[key]]$section %||% section
         cases[[key]]$subset <- cases[[key]]$subset %||% subset
+        cases[[key]]$upset_ymax <- cases[[key]]$upset_ymax %||% upset_ymax
         cases[[key]]$upset_trans <- cases[[key]]$upset_trans %||% upset_trans
     }
 }
@@ -320,7 +323,7 @@ plot_venndg <- function(counts, groups, singletons) {
     venn_p
 }
 
-plot_upset <- function(counts, singletons, upset_trans) {
+plot_upset <- function(counts, singletons, upset_ymax, upset_trans) {
 
     cnts <- column_to_rownames(counts, "CDR3.aa") %>%
         mutate(across(everything(), ~ as.integer(as.logical(.x))))
@@ -345,12 +348,21 @@ plot_upset <- function(counts, singletons, upset_trans) {
             geom_text(
                 aes(label = ..count.., vjust = ifelse(..type == "Multiplets", -0.25, +1.25)),
                 stat = "count", position = "stack", size = 2.8)
+        if (!is.null(upset_ymax)) {
+            p <- p + ylim(0, upset_ymax)
+        }
     } else {
         p <- p + geom_bar(stat = "count", position = "dodge2") +
             geom_text(
                 aes(label = ..count..),
-                stat = "count", position = position_dodge(width = 0.9), vjust = -0.25, size = 2.5) +
-            scale_y_continuous(trans = "log10")
+                stat = "count", position = position_dodge(width = 0.9), vjust = -0.25, size = 2.5)
+
+        # limit the y and do log10 transformation
+        if (!is.null(upset_ymax)) {
+            p <- p + scale_y_continuous(trans = "log10", limits = c(1, upset_ymax))
+        } else {
+            p <- p + scale_y_continuous(trans = "log10")
+        }
     }
 
     upset(
@@ -519,7 +531,7 @@ handle_subject <- function(i, subjects, casename, case) {
     upset_dir <- file.path(casedir, "upset")
     upset_png <- file.path(upset_dir, paste0("upset_", slugify(subject), ".png"))
     png(upset_png, res = 100, height = 600, width = 800)
-    print(plot_upset(counts, singletons, case$upset_trans))
+    print(plot_upset(counts, singletons, case$upset_ymax, case$upset_trans))
     dev.off()
 
     h <- headings(case$section, casename, "Overlapping Clones (UpSet Plots)")