biopipen 0.32.3__py3-none-any.whl → 0.33.0__py3-none-any.whl

biopipen/scripts/vcf/BcftoolsAnnotate.py

@@ -6,11 +6,11 @@ from biopipen.utils.reference import tabix_index
 from biopipen.utils.misc import logger
 from biopipen.scripts.vcf.bcftools_utils import run_bcftools
 
-infile = {{in.infile | repr}}  # pyright: ignore # noqa: E999
-annfile = {{in.annfile | repr}}  # pyright: ignore
-outfile = {{out.outfile | repr}}  # pyright: ignore
-joboutdir = {{job.outdir | repr}}  # pyright: ignore
-envs = {{envs | dict | repr}}  # pyright: ignore
+infile: str = {{in.infile | quote}}  # pyright: ignore # noqa: E999
+annfile: str = {{in.annfile | quote}}  # pyright: ignore
+outfile: str = {{out.outfile | quote}}  # pyright: ignore
+joboutdir: str = {{job.outdir | quote}}  # pyright: ignore
+envs: dict = {{envs | dict | repr}}  # pyright: ignore
 
 bcftools = envs.pop("bcftools")
 tabix = envs.pop("tabix")
@@ -25,14 +25,14 @@ if isinstance(columns, list):
     columns = ",".join(columns)
 
 if "c" in envs:
-    logger.warning("Ignoring envs\[c], use envs\[columns] instead.")
+    logger.warning(r"Ignoring envs\[c], use envs\[columns] instead.")
     del envs["c"]
 
 if isinstance(remove, list):
     remove = ",".join(remove)
 
 if "x" in envs:
-    logger.warning("Ignoring envs\[x], use envs\[remove] instead.")
+    logger.warning(r"Ignoring envs\[x], use envs\[remove] instead.")
     del envs["x"]
 
 envs_has_annfile = "a" in envs or "annotations" in envs
@@ -43,7 +43,7 @@ if header:
 
 if annfile and envs_has_annfile:
     logger.warning(
-        "Ignoring envs\[a/annotations] because in.annfile is provided."
+        r"Ignoring envs\[a/annotations] because in.annfile is provided."
     )
     with suppress(KeyError):
         del envs["a"]

biopipen/scripts/vcf/BcftoolsFilter.py

@@ -3,11 +3,11 @@ from pathlib import Path, PosixPath  # noqa: F401
 from biopipen.utils.misc import logger
 from biopipen.scripts.vcf.bcftools_utils import run_bcftools
 
-infile = {{in.infile | repr}}  # pyright: ignore # noqa: #999
-outfile = {{out.outfile | repr}}  # pyright: ignore
+infile: str | Path = {{in.infile | quote}}  # pyright: ignore # noqa: #999
+outfile: str = {{out.outfile | quote}}  # pyright: ignore
 outdir = Path(outfile).parent
 
-envs = {{envs | dict | repr}}  # pyright: ignore
+envs: dict = {{envs | dict | repr}}  # pyright: ignore
 bcftools = envs.pop("bcftools")
 tabix = envs.pop("tabix")
 keep = envs.pop("keep")

biopipen/scripts/vcf/BcftoolsMerge.py

@@ -0,0 +1,31 @@
+from biopipen.utils.reference import tabix_index
+from biopipen.utils.misc import logger
+from biopipen.scripts.vcf.bcftools_utils import run_bcftools
+
+infiles: list = {{in.infiles | each: as_path}}  # pyright: ignore # noqa: E999
+outfile = {{out.outfile | repr}}  # pyright: ignore
+joboutdir = {{job.outdir | repr}}  # pyright: ignore
+envs: dict = {{envs | dict | repr}}  # pyright: ignore
+
+bcftools = envs.pop("bcftools")
+tabix = envs.pop("tabix")
+ncores = envs.pop("ncores")
+gz = envs.pop("gz")
+index = envs.pop("index")
+
+envs.setdefault("force-single", True)
+envs.setdefault("missing-to-ref", True)
+
+if index and not gz:
+    logger.warning("Forcing envs.gz to True because envs.index is True.")
+    gz = True
+
+if "O" not in envs and "output-type" not in envs and "output_type" not in envs:
+    envs["O"] = "z" if gz else "v"
+
+envs[""] = [bcftools, "merge"]
+envs["o"] = outfile
+envs["threads"] = ncores
+envs["_"] = infiles
+
+run_bcftools(envs, bcftools=bcftools, index=index, tabix=tabix)

biopipen/scripts/vcf/BcftoolsSort.py

@@ -4,9 +4,9 @@ from pathlib import Path, PosixPath  # noqa: F401
 from biopipen.utils.misc import run_command, logger
 from biopipen.scripts.vcf.bcftools_utils import run_bcftools
 
-infile = {{in.infile | quote}}  # pyright: ignore # noqa: E999
-outfile = {{out.outfile | quote}}  # pyright: ignore
-envs = {{envs | dict | repr}}  # pyright: ignore
+infile: str = {{in.infile | quote}}  # pyright: ignore # noqa: E999
+outfile: str = {{out.outfile | quote}}  # pyright: ignore
+envs: dict = {{envs | dict | repr}}  # pyright: ignore
 
 outdir = Path(outfile).parent
 bcftools = envs.pop("bcftools")
@@ -97,7 +97,7 @@ if chrsize:
         infile
     ], fg=True)
 
-    infile = reheader_vcf
+    infile = str(reheader_vcf)
 
 envs[""] = [bcftools, "sort"]
 envs["_"] = infile

biopipen/scripts/vcf/BcftoolsView.py

@@ -6,10 +6,10 @@ from biopipen.utils.misc import logger
 from biopipen.utils.reference import tabix_index
 from biopipen.scripts.vcf.bcftools_utils import run_bcftools
 
-infile = {{in.infile | repr}}  # pyright: ignore # noqa: #999
-regions_file = {{in.regions_file | repr}}  # pyright: ignore
-samples_file = {{in.samples_file | repr}}  # pyright: ignore
-outfile = {{out.outfile | repr}}  # pyright: ignore
+infile: str = {{in.infile | quote}}  # pyright: ignore # noqa: #999
+regions_file: str = {{in.regions_file | quote}}  # pyright: ignore
+samples_file: str = {{in.samples_file | quote}}  # pyright: ignore
+outfile: str = {{out.outfile | quote}}  # pyright: ignore
 envs: dict = {{envs | dict | repr}}  # pyright: ignore
 
 bcftools = envs.pop("bcftools")
@@ -21,7 +21,7 @@ index = envs.pop("index")
 if regions_file:
     if "R" in envs or "regions_file" in envs or "regions-file" in envs:
         logger.warning(
-            "Ignoring envs\[regions_file/regions-file/R] "
+            r"Ignoring envs\[regions_file/regions-file/R] "
             "because in.regionsfile is provided."
         )
         with suppress(KeyError):

biopipen/scripts/vcf/Vcf2Bed.py

@@ -1,7 +1,7 @@
 from cyvcf2 import VCF, Variant
 
-infile = {{in.infile | quote}}  # pyright: ignore
-outfile = {{out.outfile | quote}}  # pyright: ignore
+infile: str = {{in.infile | quote}}  # pyright: ignore  # noqa: E999
+outfile: str = {{out.outfile | quote}}  # pyright: ignore
 # vcf, default 1
 inbase = {{envs.inbase | int}}  # pyright: ignore
 # bed, default 0

biopipen/scripts/vcf/VcfAnno.py

@@ -2,22 +2,22 @@ from os import path
 
 from biopipen.utils.misc import run_command, dict_to_cli_args
 
-infile = {{in.infile | quote}}  # pyright: ignore
-outfile = {{out.outfile | quote}}  # pyright: ignore
-joboutdir = {{job.outdir | quote}}  # pyright: ignore
+infile: str = {{in.infile | quote}}  # pyright: ignore  # noqa
+outfile: str = {{out.outfile | quote}}  # pyright: ignore
+joboutdir: str = {{job.outdir | quote}}  # pyright: ignore
 vcfanno = {{envs.vcfanno | quote}}  # pyright: ignore
-ncores = {{envs.ncores | repr}}  # pyright: ignore
-args = {{envs.args | repr}}  # pyright: ignore
+ncores: int = {{envs.ncores | repr}}  # pyright: ignore
+args: dict = {{envs.args | dict}}  # pyright: ignore
 
-{% set conf = envs.conffile or in.conffile %}
-{% if conf | isinstance: dict %}
+{% set conf = envs.conffile or in.conffile %}  # pyright: ignore  # noqa
+{% if conf | isinstance: dict %}  # pyright: ignore  # noqa
 conffile = path.join(joboutdir, "config.toml")
-conf = {{ conf | toml | quote }}
+conf: str = {{ conf | toml | quote }}  # pyright: ignore  # noqa
 with open(conffile, "w") as f:
     f.write(conf)
-{% else %}
-conffile = {{conf | quote}}
-{% endif %}
+{% else %}  # pyright: ignore  # noqa
+conffile = {{conf | quote}}  # pyright: ignore  # noqa
+{% endif %}  # pyright: ignore  # noqa
 
 args["p"] = ncores
 args["_"] = [conffile, infile]

biopipen/scripts/vcf/VcfDownSample.sh

@@ -1,25 +1,37 @@
+# shellcheck disable=SC2148
+# shellcheck disable=SC2036
+# shellcheck disable=SC2030
+# shellcheck disable=SC1083
+# shellcheck disable=SC2288
 infile={{in.infile | quote}}
 outfile={{out.outfile | quote}}
 n={{envs.n}}
 
+# shellcheck disable=SC2031
 if [[ $infile == *.gz ]]; then
-    outfile=$(echo $outfile | sed -r "s/\.gz$//")
-    nheader=$(zcat $infile | head -n 9999 | grep "^#" | wc -l | cut -d' ' -f1)
+    outfile=$(echo "$outfile" | sed -r "s/\.gz$//")
+    # shellcheck disable=SC2126
+    nheader=$(zcat "$infile" | head -n 9999 | grep "^#" | wc -l | cut -d' ' -f1)
     if [[ ! $n -gt 1 ]]; then
-        nrows=$(zcat $infile | wc -l | cut -d' ' -f1)
+        nrows=$(zcat "$infile" | wc -l | cut -d' ' -f1)
+        # shellcheck disable=SC2004
         nvars=$(($nrows - $nheader))
         n=$(echo "$nvars * $n" | bc)
     fi
-    zcat $infile | head -n $nheader > $outfile
-    zcat $infile | tail -n +$(($nheader + 1)) | shuf -n $n | LC_ALL=C sort -k1,1V -k2,2n >> $outfile
-    bgzip $outfile
+    zcat "$infile" | head -n "$nheader" > "$outfile"
+    # shellcheck disable=SC2004
+    zcat "$infile" | tail -n +$(($nheader + 1)) | shuf -n "$n" | LC_ALL=C sort -k1,1V -k2,2n >> "$outfile"
+    bgzip "$outfile"
 else
-    nheader=$(head -n 9999 $infile | grep "^#" | wc -l | cut -d' ' -f1)
+    # shellcheck disable=SC2126
+    nheader=$(head -n 9999 "$infile" | grep "^#" | wc -l | cut -d' ' -f1)
     if [[ ! $n -gt 1 ]]; then
-        nrows=$(wc -l $infile | cut -d' ' -f1)
+        nrows=$(wc -l "$infile" | cut -d' ' -f1)
+        # shellcheck disable=SC2004
         nvars=$(($nrows - $nheader))
         n=$(echo "$nvars * $n" | bc)
     fi
-    head -n $nheader $infile > $outfile
-    tail -n +$(($nheader + 1)) $infile | shuf -n $n | LC_ALL=C sort -k1,1V -k2,2n >> $outfile
+    head -n "$nheader" "$infile" > "$outfile"
+    # shellcheck disable=SC2004
+    tail -n +$(($nheader + 1)) "$infile" | shuf -n "$n" | LC_ALL=C sort -k1,1V -k2,2n >> "$outfile"
 fi

biopipen/scripts/vcf/VcfFilter.py

@@ -1,13 +1,13 @@
 from cyvcf2 import VCF, Writer, Variant
 
-infile = {{in.invcf | repr}}  # pyright: ignore
-outfile = {{out.outfile | repr}}  # pyright: ignore
+infile: str = {{in.invcf | quote}}  # pyright: ignore  # noqa: E999
+outfile: str = {{out.outfile | quote}}  # pyright: ignore
 
-{{envs.helper}}
+{{envs.helper}}  # pyright: ignore  # noqa: E999
 
 keep = {{envs.keep | repr}}  # pyright: ignore
-filters = {{envs.filters | repr}}  # pyright: ignore
-filter_descs = {{envs.filter_descs | repr}}  # pyright: ignore
+filters: dict = {{envs.filters | repr}}  # pyright: ignore
+filter_descs: dict = {{envs.filter_descs | repr}}  # pyright: ignore
 
 # builtin filters
 BUILTIN_FILTERS = {}

biopipen/scripts/vcf/VcfFix.py

@@ -7,17 +7,17 @@ from biopipen.scripts.vcf.VcfFix_utils import (  # noqa: F401
     HeaderContig,
     HeaderGeneral,
     Fields,
-    Info,
-    Format,
-    Alt,
-    Filter,
-    Sample,
-    Samples,
+    # Info,
+    # Format,
+    # Alt,
+    # Filter,
+    # Sample,
+    # Samples,
     Variant,
 )
 from biopipen.scripts.vcf.VcfFix_utils import fix_vcffile
 
-infile = {{in.infile | quote}}  # pyright: ignore
+infile = {{in.infile | quote}}  # pyright: ignore  # noqa: E999
 instem = {{in.infile | stem | quote}}  # pyright: ignore
 outfile = {{out.outfile | quote}}  # pyright: ignore
 

biopipen/scripts/vcf/VcfFix_utils.py

@@ -1,6 +1,15 @@
 import re
 import gzip
-from biopipen.utils.vcf import *  # noqa: F401, F403
+from biopipen.utils.vcf import (
+    HeaderInfo,
+    HeaderFormat,
+    HeaderFilter,
+    HeaderContig,
+    HeaderGeneral,
+    Fields,
+    Variant,
+    HeaderItem,
+)
 
 
 def line_to_obj(line: str):
@@ -41,7 +50,7 @@ def handle_obj(obj, fixes: dict):
 
         regex = fix.get("regex")
         if regex:
-            if not re.search(regex, obj.raw):
+            if not re.search(regex, obj.raw):  # type: ignore
                 continue
 
             return fix["fix"](obj.raw if kind is None else obj)
@@ -67,7 +76,7 @@ def fix_vcffile(vcffile, outfile, fixes):
     with inopen(vcffile, "rt") as fin, open(outfile, "w") as fout:
         for line in fin:
             obj = line_to_obj(line)
-            out = handle_obj(obj, modify_fixes)
+            out = handle_obj(obj, modify_fixes)  # type: ignore
             if obj.kind == "fields":
                 for fix in header_append_fixes:
                     fout.write(str(fix["fix"](None)).rstrip("\n") + "\n")

biopipen/scripts/vcf/VcfIndex.py

@@ -4,10 +4,10 @@ from os import path
 from biopipen.utils.reference import tabix_index
 from biopipen.utils.misc import run_command
 
-infile = {{in.infile | repr}}  # pyright: ignore
-outfile = Path({{out.outfile | repr}})  # pyright: ignore
+infile: str = {{in.infile | quote}}  # pyright: ignore  # noqa
+outfile = Path({{out.outfile | quote}})  # pyright: ignore
 outidx = {{out.outidx | repr}}  # pyright: ignore
-tabix = {{envs.tabix | repr}}  # pyright: ignore
+tabix: str = {{envs.tabix | repr}}  # pyright: ignore
 ncores = {{envs.ncores | repr}}  # pyright: ignore
 
 outfile_with_index = tabix_index(infile, "vcf", outfile.parent, tabix)

biopipen/scripts/vcf/VcfIntersect.py

@@ -1,8 +1,8 @@
 from biopipen.utils.misc import run_command, dict_to_cli_args
 
-infile1 = {{in.infile1 | repr}}  # pyright: ignore
-infile2 = {{in.infile2 | repr}}  # pyright: ignore
-outfile = {{out.outfile | repr}}  # pyright: ignore
+infile1: str = {{in.infile1 | quote}}  # pyright: ignore  # noqa
+infile2 = {{in.infile2 | quote}}  # pyright: ignore
+outfile = {{out.outfile | quote}}  # pyright: ignore
 bcftools = {{envs.bcftools | repr}}  # pyright: ignore
 gz = {{envs.gz | repr}}  # pyright: ignore
 index = {{envs.index | repr}}  # pyright: ignore

biopipen/scripts/vcf/VcfLiftOver.sh

@@ -1,3 +1,5 @@
+# shellcheck disable=SC2148
+# shellcheck disable=SC1083
 invcf={{ in.invcf | quote }}
 outvcf={{ out.outvcf | quote }}
 rejfile={{ job.outdir | joinpaths: "rejected.vcf" | quote }}
@@ -6,12 +8,15 @@ chain={{ envs.chain | quote }}
 reffa={{ envs.reffa | quote }}
 args={{ envs.args | dict_to_cli_args: join=True }}
 
+# shellcheck disable=SC2154
 refdict="${reffa%.fa}.dict"
 if [[ ! -e "$refdict" ]]; then
     echo "Sequence dictionary does not exist: $refdict" 1>&2
     exit 1
 fi
 
+# shellcheck disable=SC2154
+# shellcheck disable=SC2086
 $gatk LiftoverVcf \
     $args \
     --INPUT "$invcf" \

biopipen/scripts/vcf/VcfSplitSamples.py

@@ -3,12 +3,12 @@ import shlex
 import concurrent.futures
 from subprocess import Popen, check_output
 
-infile = {{in.infile | repr}}  # pyright: ignore
-outdir = {{out.outdir | repr}}  # pyright: ignore
-bcftools = {{envs.bcftools | repr}}  # pyright: ignore
+infile: str = {{in.infile | quote}}  # pyright: ignore  # noqa
+outdir: str = {{out.outdir | quote}}  # pyright: ignore
+bcftools: str = {{envs.bcftools | repr}}  # pyright: ignore
 gz = {{envs.gz | repr}}  # pyright: ignore
 index = {{envs.index | repr}}  # pyright: ignore
-ncores = {{envs.ncores | int}}  # pyright: ignore
+ncores: int = {{envs.ncores | int}}  # pyright: ignore
 private = {{envs.private | repr}}  # pyright: ignore
 
 if index:

biopipen/scripts/vcf/bcftools_utils.py

@@ -15,7 +15,7 @@ def bcftools_version(bcftools: str) -> tuple[int, ...]:
     """
     bversion = (
         run_command([bcftools, "version"], stdout="return")
-        .splitlines()[0]  # bcftools 1.20
+        .splitlines()[0]  # bcftools 1.20  # type: ignore
         .replace("bcftools", "")
         .strip()  # 1.20
         .split(".")
@@ -24,8 +24,8 @@ def bcftools_version(bcftools: str) -> tuple[int, ...]:
 
 
 def run_bcftools(
-    args: dict[str, object],
-    bcftools: str,
+    args: dict,
+    bcftools: str,  # TODO: get from the first argument of args
     index: bool,
     tabix: str
 ) -> None:

biopipen/scripts/web/Download.py

@@ -2,13 +2,13 @@ from pathlib import Path
 
 from biopipen.utils.misc import run_command, dict_to_cli_args
 
-url = {{in.url | repr}}  # pyright: ignore
-outfile = Path({{out.outfile | repr}})  # pyright: ignore
+url = {{in.url | quote}}  # pyright: ignore # noqa
+outfile = Path({{out.outfile | quote}})  # pyright: ignore
 tool = {{envs.tool | repr}}  # pyright: ignore
 wget = {{envs.wget | repr}}  # pyright: ignore
 aria2c = {{envs.aria2c | repr}}  # pyright: ignore
 ncores = {{envs.ncores | repr}}  # pyright: ignore
-args = {{envs.args | dict}}  # pyright: ignore
+args: dict = {{envs.args | dict}}  # pyright: ignore
 
 if tool == "wget":
     args["_"] = url
@@ -28,4 +28,8 @@ elif tool == "aria2c":
 
 else: # use python
     import urllib
-    urllib.urlretrieve(url, outfile)
+
+    try:
+        urllib.urlretrieve(url, outfile)  # type: ignore
+    except AttributeError:
+        urllib.request.urlretrieve(url, outfile)  # type: ignore

biopipen/scripts/web/DownloadList.py

@@ -2,13 +2,13 @@ from pathlib import Path
 
 from biopipen.utils.misc import run_command, dict_to_cli_args
 
-urlfile = {{in.urlfile | repr}}  # pyright: ignore
-outdir = Path({{out.outdir | repr}})  # pyright: ignore
+urlfile: str = {{in.urlfile | quote}}  # pyright: ignore  # noqa
+outdir = Path({{out.outdir | quote}})  # pyright: ignore
 tool = {{envs.tool | repr}}  # pyright: ignore
 wget = {{envs.wget | repr}}  # pyright: ignore
 aria2c = {{envs.aria2c | repr}}  # pyright: ignore
 ncores = {{envs.ncores | repr}}  # pyright: ignore
-args = {{envs.args | repr}}  # pyright: ignore
+args: dict = {{envs.args | repr}}  # pyright: ignore
 
 if tool == "wget":
     args["i"] = urlfile
@@ -26,10 +26,10 @@ elif tool == "aria2c":
     run_command(dict_to_cli_args(args, dashify=True), fg=True)
 
 else: # use python
-    import urllib
+    from urllib.request import urlretrieve
     from urllib.parse import urlparse
     with open(urlfile, "r") as furl:
         for i, url in enumerate(furl.readlines()):
             parsed = urlparse(url)
             path = Path(parsed.path)
-            urllib.urlretrieve(url, f"{path.stem}-{i}{path.suffix}")
+            urlretrieve(url, f"{path.stem}-{i}{path.suffix}")

biopipen/scripts/web/GCloudStorageDownloadBucket.py

@@ -8,12 +8,12 @@ from biopipen.scripts.web.gcloud_common import (
     get_file_path,
 )
 
-url = {{in.url | repr}}  # pyright: ignore  # noqa: E999
+url: str = {{in.url | quote}}  # pyright: ignore  # noqa: E999
 outdir = Path({{out.outdir | repr}})  # pyright: ignore
-gcloud = {{envs.gcloud | repr}}  # pyright: ignore
+gcloud: str = {{envs.gcloud | quote}}  # pyright: ignore
 keep_structure = {{envs.keep_structure | repr}}  # pyright: ignore
-ncores = {{envs.ncores | repr}}  # pyright: ignore
-args = {{envs.args | repr}}  # pyright: ignore
+ncores: int = {{envs.ncores | repr}}  # pyright: ignore
+args: dict = {{envs.args | repr}}  # pyright: ignore
 
 if not is_valid_gs_bucket_url(url):
     raise Exception(
@@ -65,7 +65,7 @@ def download_file(i: int, line: str, total: int):
 def download_bucket():
     out = run_command([gcloud, "storage", "ls", "--recursive", url], stdout="RETURN")
     # remove empty lines and skip the root
-    out = list(filter(None, out.splitlines()[1:]))
+    out = list(filter(None, out.splitlines()[1:]))  # type: ignore
     if keep_structure:
         # create folders first
         logger.info(f"Creating folders to keep structure.")

biopipen/scripts/web/GCloudStorageDownloadFile.py

@@ -1,10 +1,10 @@
 from biopipen.utils.misc import run_command, dict_to_cli_args
 from biopipen.scripts.web.gcloud_common import is_logged_in, is_valid_gs_file_url
 
-url = {{in.url | repr}}  # pyright: ignore  # noqa: E999
+url: str = {{in.url | repr}}  # pyright: ignore  # noqa: E999
 outfile = {{out.outfile | repr}}  # pyright: ignore
-gcloud = {{envs.gcloud | repr}}  # pyright: ignore
-args = {{envs.args | repr}}  # pyright: ignore
+gcloud: str = {{envs.gcloud | repr}}  # pyright: ignore
+args: dict = {{envs.args | repr}}  # pyright: ignore
 
 if not is_valid_gs_file_url(url):
     raise Exception(

biopipen/scripts/web/gcloud_common.py

@@ -12,7 +12,7 @@ def is_logged_in(gcloud: str) -> bool:
         bool: True if the user is logged in, False otherwise.
     """
     out = run_command([gcloud, "auth", "list"], stdout="RETURN")
-    return "ACTIVE" in out
+    return "ACTIVE" in out  # type: ignore
 
 
 def is_valid_gs_bucket_url(url: str) -> bool:

biopipen/utils/gsea.R

@@ -34,46 +34,85 @@ if (!exists("slugify")) {
     }
 }
 
+#' Download the GMT file and save it to cachedir
+#' Return the path to the GMT file
+#' We also check if the second column is shorter than the first column.
+#' If so, we switch the first and second columns.
+#' In case some providers provide the GMT file with the first and second columns switched.
+#' We also replace the "/" in the gene names with "-" if any. This is because the "/" is
+#' not allowed in a path, but GSEA uses the gene names as the file name.
+#'
+#' @param gmturl The URL or path of the GMT file
+#' @param cachedir The directory to save the GMT file
+#' @return The path to the GMT file
 localizeGmtfile <- function(gmturl, cachedir = tempdir()) {
     # Download the GMT file and save it to cachedir
     # Return the path to the GMT file
-    if (!startsWith(gmturl, "http") && !startsWith(gmturl, "ftp")) {
-        return(gmturl)
+    in_gmtfile <- out_gmtfile <- file.path(cachedir, basename(gmturl))
+    if (startsWith(gmturl, "http") || startsWith(gmturl, "ftp")) {
+        download.file(gmturl, in_gmtfile)
+        remote <- TRUE
+    } else {
+        in_gmtfile <- gmturl
+        remote <- FALSE
     }
-    gmtfile = file.path(cachedir, basename(gmturl))
-    if (!file.exists(gmtfile)) {
-        download.file(gmturl, gmtfile)
-        items <- read.delim(gmtfile, header = FALSE, stringsAsFactors = FALSE, sep = "\t")
-        if (ncol(items) < 3) {
-            stop(paste0("Invalid GMT file: ", gmtfile, ", from ", gmturl))
-        }
-        if (nrow(items) == 0) {
-            stop(paste0("Empty GMT file: ", gmtfile, ", from ", gmturl))
-        }
-        if (
-            is.character(items$V2[1]) &&
-            nchar(items$V2[1]) < nchar(items$V1[1]) &&
-            nchar(items$V2[1]) > 0 &&
-            is.na(suppressWarnings(as.numeric(items$V2[1])))
-        ) {
-            warning(paste0(
-                "The second column is shorter, switching the first and second columns in GMT file ",
-                gmtfile,
-                " from ",
-                gmturl
-            ))
-            items <- items[, c(2, 1, 3:ncol(items))]
-            write.table(
-                items,
-                gmtfile,
-                row.names = F,
-                col.names = F,
-                sep = "\t",
-                quote = F
-            )
-        }
+
+    items <- readLines(in_gmtfile)
+    items <- items[!grepl("^#", items) & nchar(items) > 0]
+    items <- lapply(strsplit(items, "\t"), function(x) c(x[1:2], paste0(x[3:length(x)], collapse = "\t")))
+    items <- as.data.frame(t(as.data.frame(items)))
+    rownames(items) <- NULL
+    colnames(items) <- c("V1", "V2", "V3")
+
+    if (ncol(items) < 3) {
+        stop(paste0("Invalid GMT file: ", gmturl))
+    }
+    if (nrow(items) == 0) {
+        stop(paste0("Empty GMT file: ", gmturl))
     }
-    return(gmtfile)
+
+    # Check if the second column is shorter than the first column
+    nchars1 <- sum(nchar(items$V1))
+    nchars2 <- sum(nchar(items$V2))
+    prefix <- gsub("[0-9]+$", "", items$V2[1])
+
+    if (is.character(items$V2) &&     # Only when V2 is character, as pathway names
+        nchars2 < nchars1 &&          # Only when V2 is shorter than V1
+        all(nchar(items$V2) > 0) &&   # Only when V2 is not empty
+        !all(grepl("^[0-9]+$", items$V2)) && # Only when V2 is not all numbers
+        (nchar(prefix) == 0 || !all(startsWith(items$V2, prefix))) # Only when they are not like hsa00001, hsa00002, etc.
+    ) {
+        warning(paste0(
+            "The second column is shorter, switching the first and second columns in ",
+            "GMT file ", gmturl
+        ))
+        items <- items[, c(2, 1, 3:ncol(items))]
+        switched <- TRUE
+    } else {
+        switched <- FALSE
+    }
+
+    if (any(grepl("/", items$V1))) {
+        items$V1 <- gsub("/", "-", items$V1)
+        replaced <- TRUE
+    } else {
+        replaced <- FALSE
+    }
+
+    if (remote || switched || replaced) {
+        write.table(
+            items,
+            out_gmtfile,
+            row.names = FALSE,
+            col.names = FALSE,
+            sep = "\t",
+            quote = FALSE
+        )
+    } else {
+        out_gmtfile <- in_gmtfile
+    }
+
+    return(out_gmtfile)
 }
 
 
@@ -261,6 +300,7 @@ runGSEA = function(
         mutate(Description = "na") %>%
         rownames_to_column("NAME") %>%
         select(NAME, Description, everything())
+
     write.table(
         indata,
         gctfile,