biopipen 0.31.4__py3-none-any.whl → 0.31.6__py3-none-any.whl

biopipen/__init__.py

@@ -1 +1 @@
-__version__ = "0.31.4"
+__version__ = "0.31.6"

biopipen/ns/bam.py

@@ -260,3 +260,44 @@ class BamMerge(Proc):
         "sort_args": [],
     }
     script = "file://../scripts/bam/BamMerge.py"
+
+
+class BamSampling(Proc):
+    """Keeping only a fraction of read pairs from a bam file
+
+    Input:
+        bamfile: The bam file
+
+    Output:
+        outfile: The output bam file
+
+    Envs:
+        ncores: Number of cores to use
+        samtools: Path to samtools executable
+        tool: The tool to use, currently only "samtools" is supported
+        fraction (type=float): The fraction of reads to keep.
+            If `0 < fraction <= 1`, it's the fraction of reads to keep.
+            If `fraction > 1`, it's the number of reads to keep.
+            Note that when fraction > 1, you may not get the exact number
+            of reads specified but a close number.
+        seed: The seed for random number generator
+        index: Whether to index the output bam file
+        sort: Whether to sort the output bam file
+        sort_args: The arguments for sorting bam file using `samtools sort`.
+            These keys are not allowed: `-o`, `-@`,
+            and `--threads`, as they are managed by the script.
+    """
+    input = "bamfile:file"
+    output = "outfile:file:{{in.bamfile | stem}}.sampled{{envs.fraction}}.bam"
+    lang = config.lang.python
+    envs = {
+        "ncores": config.misc.ncores,
+        "samtools": config.exe.samtools,
+        "tool": "samtools",
+        "fraction": None,
+        "seed": 8525,
+        "index": True,
+        "sort": True,
+        "sort_args": [],
+    }
+    script = "file://../scripts/bam/BamSampling.py"

biopipen/ns/protein.py

@@ -0,0 +1,84 @@
+"""Protein-related processes."""
+from ..core.proc import Proc
+from ..core.config import config
+
+
+class Prodigy(Proc):
+    """Prediction of binding affinity of protein-protein complexes based on
+    intermolecular contacts using Prodigy.
+
+    See <https://rascar.science.uu.nl/prodigy/> and
+    <https://github.com/haddocking/prodigy>.
+
+    `prodigy-prot` must be installed under the given python of `proc.lang`.
+
+    Input:
+        infile: The structure file in PDB or mmCIF format.
+
+    Output:
+        outfile: The output file generated by Prodigy.
+        outdir: The output directory containing all output files.
+
+    Envs:
+        distance_cutoff (type=float): The distance cutoff to calculate intermolecular
+            contacts.
+        acc_threshold (type=float): The accessibility threshold for BSA analysis.
+        temperature (type=float): The temperature (C) for Kd prediction.
+        contact_list (flag): Whether to generate contact list.
+        pymol_selection (flag): Whether output a script to highlight the interface
+            residues in PyMOL.
+        selection (list): The selection of the chains to analyze.
+            `['A', 'B']` will analyze chains A and B.
+            `['A,B', 'C']` will analyze chain A and C; and B and C.
+            `['A', 'B', 'C']` will analyze all combinations of A, B, and C.
+        outtype (choice): Set the format of the output file (`out.outfile`).
+            All three files will be generated. This option only determines which
+            is assigned to `out.outfile`.
+            - raw: The raw output file from prodigy.
+            - json: The output file in JSON format.
+            - tsv: The output file in CSV format.
+    """
+    input = "infile:file"
+    output = [
+        "outfile:file:{{in.infile | stem}}_prodigy/"
+        "{{in.infile | stem}}.{{envs.outtype if envs.outtype != 'raw' else 'out'}}",
+        "outdir:dir:{{in.infile | stem}}_prodigy",
+    ]
+    lang = config.lang.python
+    envs = {
+        "distance_cutoff": 5.5,
+        "acc_threshold": 0.05,
+        "temperature": 25.0,
+        "contact_list": True,
+        "pymol_selection": True,
+        "selection": None,
+        "outtype": "json",
+    }
+    script = "file://../scripts/protein/Prodigy.py"
+
+
+class ProdigySummary(Proc):
+    """Summary of the output from `Prodigy`.
+
+    Input:
+        infiles: The output json file generated by `Prodigy`.
+
+    Output:
+        outdir: The directory of summary files generated by `ProdigySummary`.
+
+    Envs:
+        group (type=auto): The group of the samples for boxplots.
+            If `None`, don't do boxplots.
+            It can be a dict of group names and sample names, e.g.
+            `{"group1": ["sample1", "sample2"], "group2": ["sample3"]}`
+            or a file containing the group information, with the first column
+            being the sample names and the second column being the group names.
+            The file should be tab-delimited with no header.
+    """
+    input = "infiles:files"
+    input_data = lambda ch: [[f"{odir}/_prodigy.tsv" for odir in ch.outdir]]
+    output = "outdir:dir:prodigy_summary"
+    lang = config.lang.rscript
+    envs = {"group": None}
+    script = "file://../scripts/protein/ProdigySummary.R"
+    plugin_opts = {"report": "file://../reports/protein/ProdigySummary.svelte"}

biopipen/ns/regulatory.py

@@ -212,3 +212,75 @@ class MotifAffinityTest(Proc):
         "atsnp_args": {"padj_cutoff": True, "padj": "BH", "p": "pval_diff"},
     }
     script = "file://../scripts/regulatory/MotifAffinityTest.R"
+
+
+class VariantMotifPlot(Proc):
+    """A plot with a genomic region surrounding a genomic variant, and
+    potentially disrupted motifs.
+
+    Currently only SNVs are supported.
+
+    Input:
+        infile: File containing the variants and motifs.
+            It is a TAB-delimited file with the following columns:
+            - chrom: The chromosome of the SNV. Alias: chr, seqnames.
+            - start: The start position of the SNV, no matter 0- or 1-based.
+            - end: The end position of the SNV, which will be used as the position of the SNV.
+            - strand: Indicating the direction of the surrounding sequence matching the motif.
+            - SNP_id: The name of the SNV.
+            - REF: The reference allele of the SNV.
+            - ALT: The alternative allele of the SNV.
+            - providerId: The motif id. It can be specified by `envs.motif_col`.
+            - providerName: The name of the motif provider. Optional.
+            - Regulator: The regulator name. Optional, can be specified by `envs.regulator_col`.
+            - motifPos: The position of the motif, relative to the position of the SNV.
+                For example, '-8, 4' means the motif is 8 bp upstream and 4 bp downstream of the SNV.
+
+    Envs:
+        genome: The genome assembly.
+            Used to fetch the sequences around the variants by package, for example, `BSgenome.Hsapiens.UCSC.hg19` is required if
+            `hg19`. If it is an organism other than human, please specify the full name of the package, for example, `BSgenome.Mmusculus.UCSC.mm10`.
+        motifdb: The path to the motif database. This is required.
+            It should be in the format of MEME motif database.
+            Databases can be downloaded here: <https://meme-suite.org/meme/doc/download.html>.
+            See also introduction to the databases: <https://meme-suite.org/meme/db/motifs>.
+            [universalmotif](https://github.com/bjmt/universalmotif) is required to read the motif database.
+        motif_col: The column name in the motif file containing the motif names.
+            If this is not provided, `envs.regulator_col` and `envs.regmotifs` are required,
+            which are used to infer the motif names from the regulator names.
+        regulator_col: The column name in the motif file containing the regulator names.
+            Both `motif_col` and `regulator_col` should be the direct column names or
+            the index (1-based) of the columns.
+            If no `regulator_col` is provided, no regulator information is written in
+            the output. Otherwise, the regulator information is written in the output in
+            the `Regulator` column.
+        regmotifs: The path to the regulator-motif mapping file.
+            It must have header and the columns `Motif` or `Model` for motif names and
+            `TF`, `Regulator` or `Transcription factor`  for regulator names.
+        notfound (choice): What to do if a motif is not found in the database,
+            or a regulator is not found in the regulator-motif mapping (envs.regmotifs)
+            file.
+            - error: Report error and stop the process.
+            - ignore: Ignore the motif and continue.
+        devpars (ns): The default device parameters for the plot.
+            - width (type=int): The width of the plot.
+            - height (type=int): The height of the plot.
+            - res (type=int): The resolution of the plot.
+        plot_vars (type=auto): The variants (SNP_id) to plot.
+            A list of variant names to plot or a string with the variant names separated by comma.
+            When not specified, all variants are plotted.
+    """  # noqa: E501
+    input = "infile:file"
+    output = "outdir:dir:{{in.infile | stem}}.vmplots"
+    lang = config.lang.rscript
+    envs = {
+        "genome": config.ref.genome,
+        "motifdb": config.ref.tf_motifdb,
+        "motif_col": "providerId",
+        "regulator_col": None,
+        "regmotifs": config.ref.tf_motifs,
+        "notfound": "error",
+        "devpars": {"width": 800, "height": None, "res": 100},
+        "plot_vars": None,
+    }
+    script = "file://../scripts/regulatory/VariantMotifPlot.R"

biopipen/ns/vcf.py

@@ -335,6 +335,8 @@ class TruvariBench(Proc):
     """Run `truvari bench` to compare a VCF with CNV calls and
     base CNV standards
 
+    Requires truvari v4+
+
     See https://github.com/ACEnglish/truvari/wiki/bench
 
     Input:
@@ -358,7 +360,7 @@ class TruvariBench(Proc):
         "truvari": config.exe.truvari,
         "ref": config.ref.reffa,
         "refdist": 500,
-        "pctsim": 0.7,
+        "pctseq": 0.7,
         "pctsize": 0.7,
         "pctovl": 0.0,
         "typeignore": False,
@@ -402,7 +404,7 @@ class TruvariBenchSummary(Proc):
     output = "outdir:dir:truvari_bench.summary"
     lang = config.lang.rscript
     envs = {
-        "plots": ["call cnt", "base cnt", "precision", "recall", "f1"],
+        "plots": ["comp cnt", "base cnt", "precision", "recall", "f1"],
         "devpars": None,
     }
     script = "file://../scripts/vcf/TruvariBenchSummary.R"
@@ -414,6 +416,8 @@ class TruvariConsistency(Proc):
 
     See https://github.com/ACEnglish/truvari/wiki/consistency
 
+    Requires truvari v4+
+
     Input:
         vcfs: The vcf files with CNV calls
 
@@ -463,7 +467,7 @@ class BcftoolsAnnotate(Proc):
         columns (auto): Comma-separated or list of columns or tags to carry over from
             the annotation file. Overrides `-c, --columns`
         remove (auto): Remove the specified columns from the input file
-        header (type=list): Headers to be added
+        header (list): Headers to be added
         gz (flag): Whether to gzip the output file
         index (flag): Whether to index the output file (tbi) (`envs.gz` forced to True)
         <more>: Other arguments for `bcftools annotate`

biopipen/reports/protein/ProdigySummary.svelte

@@ -0,0 +1,16 @@
+{% from "utils/misc.liq" import report_jobs -%}
+
+<script>
+    import { Image, DataTable, Descr } from "$libs";
+</script>
+
+{%- macro report_job(job, h=1) -%}
+    {{ job | render_job: h=h }}
+{%- endmacro -%}
+
+
+{%- macro head_job(job) -%}
+    <h1>{{job.out.outdir | stem | escape}}</h1>
+{%- endmacro -%}
+
+{{ report_jobs(jobs, head_job, report_job) }}

biopipen/scripts/bam/BamMerge.py

@@ -1,7 +1,7 @@
 from pathlib import Path
-from biopipen.utils.misc import run_command
+from biopipen.utils.misc import run_command, logger
 
-bamfiles = {{in.bamfiles | repr}}  # pyright: ignore
+bamfiles = {{in.bamfiles | repr}}  # pyright: ignore # noqa
 outfile = Path({{out.outfile | repr}})  # pyright: ignore
 ncores = {{envs.ncores | int}}  # pyright: ignore
 tool = {{envs.tool | quote}}  # pyright: ignore
@@ -18,7 +18,7 @@ if should_index and not should_sort:
 
 def use_samtools():
     """Use samtools to merge bam files"""
-    print("Using samtools")
+    logger.info("Using samtools ...")
     ofile = (
         outfile
         if not should_sort
@@ -43,11 +43,11 @@ def use_samtools():
         *merge_args,
         *bamfiles,
     ]
-    print("- Merging")
+    logger.info("- Merging the bam files ...")
     run_command(cmd)
 
     if should_sort:
-        print("- Sorting")
+        logger.info("- Sorting the merged bam file ...")
         for key in ["-o", "-@", "--threads"]:
             if key in sort_args:
                 raise ValueError(
@@ -67,16 +67,14 @@ def use_samtools():
         run_command(cmd)
 
     if should_index:
-        print("- Indexing")
+        logger.info("- Indexing the output bam file ...")
         cmd = [samtools, "index", "-@", ncores, outfile]
         run_command(cmd)
 
-    print("Done")
-
 
 def use_sambamba():
     """Use sambamba to merge bam files"""
-    print("Using sambamba")
+    logger.info("Using sambamba ...")
     ofile = (
         outfile
         if not should_sort
@@ -90,11 +88,11 @@ def use_sambamba():
             )
 
     cmd = [sambamba, "merge", "-t", ncores, *merge_args, ofile, *bamfiles]
-    print("- Merging")
+    logger.info("- Merging the bam files ...")
     run_command(cmd)
 
     if should_sort:
-        print("- Sorting")
+        logger.info("- Sorting the merged bam file ...")
         for key in ["-t", "--nthreads", "-o", "--out"]:
             if key in sort_args:
                 raise ValueError(
@@ -115,12 +113,10 @@ def use_sambamba():
         run_command(cmd)
 
     if should_index:
-        print("- Indexing")
+        logger.info("- Indexing the output bam file ...")
         cmd = [sambamba, "index", "-t", ncores, outfile]
         run_command(cmd)
 
-    print("Done")
-
 
 if __name__ == "__main__":
     if tool == "samtools":

biopipen/scripts/bam/BamSampling.py

@@ -0,0 +1,90 @@
+from pathlib import Path
+from biopipen.utils.misc import run_command, logger
+
+# using:
+# samtools view --subsample 0.1 --subsample-seed 1234 --threads 4 -b -o out.bam in.bam
+
+bamfile = {{ in.bamfile | repr }} # pyright: ignore # noqa
+outfile = Path({{ out.outfile | repr }}) # pyright: ignore
+ncores = {{ envs.ncores | int }} # pyright: ignore
+samtools = {{ envs.samtools | repr }} # pyright: ignore
+tool = {{ envs.tool | repr }} # pyright: ignore
+fraction = {{ envs.fraction | repr }} # pyright: ignore
+seed = {{ envs.seed | int }} # pyright: ignore
+should_index = {{ envs.index | repr }} # pyright: ignore
+should_sort = {{ envs.sort | repr }} # pyright: ignore
+sort_args = {{ envs.sort_args | repr }} # pyright: ignore
+
+if should_index and not should_sort:
+    raise ValueError("Indexing requires sorting")
+
+if fraction is None:
+    raise ValueError("'envs.fraction' must be provided.")
+
+if tool != "samtools":
+    raise ValueError(
+        f"Tool {tool} is not supported. "
+        "Currently only samtools is supported."
+    )
+
+if fraction > 1:
+    # calculate the fraction based on the number of reads
+    logger.info("Converting fraction > 1 to a fraction of reads.")
+    cmd = [
+        samtools,
+        "view",
+        "--threads",
+        ncores,
+        "-c",
+        bamfile
+    ]
+    nreads = run_command(cmd, stdout="return").strip()
+    fraction = fraction / float(int(nreads))
+
+ofile = (
+    outfile
+    if not should_sort
+    else outfile.with_stem(f"{outfile.stem}.unsorted")
+)
+
+cmd = [
+    samtools,
+    "view",
+    "--subsample",
+    fraction,
+    "--subsample-seed",
+    seed,
+    "--threads",
+    ncores,
+    "-b",
+    "-o",
+    ofile,
+    bamfile
+]
+run_command(cmd, fg=True)
+
+if should_sort:
+    logger.info("Sorting the output bam file.")
+    for key in ["-o", "-@", "--threads"]:
+        if key in sort_args:
+            raise ValueError(
+                f"envs.sort_args cannot contain {key}, "
+                "which is managed by the script"
+            )
+
+    cmd = [
+        samtools,
+        "sort",
+        "-@",
+        ncores,
+        *sort_args,
+        "-o",
+        outfile,
+        ofile
+    ]
+    run_command(cmd, fg=True)
+
+if should_index:
+    logger.info("Indexing the output bam file.")
+    cmd = [samtools, "index", "-@", ncores, outfile]
+    run_command(cmd, fg=True)

biopipen/scripts/protein/Prodigy.py

@@ -0,0 +1,119 @@
+import json
+import logging
+import sys
+from pathlib import Path
+from prodigy_prot.predict_IC import (
+    Prodigy,
+    check_path,
+    parse_structure,
+)
+
+infile = {{in.infile | repr}}  # pyright: ignore # noqa
+outfile = {{out.outfile | repr}}  # pyright: ignore
+outdir = {{out.outdir | repr}}  # pyright: ignore
+distance_cutoff = {{envs.distance_cutoff | float}}  # pyright: ignore
+acc_threshold = {{envs.acc_threshold | float}}  # pyright: ignore
+temperature = {{envs.temperature | float}}  # pyright: ignore
+contact_list = {{envs.contact_list | repr}}  # pyright: ignore
+pymol_selection = {{envs.pymol_selection | repr}}  # pyright: ignore
+selection = {{envs.selection | repr}}  # pyright: ignore
+outtype = {{envs.outtype | repr}}  # pyright: ignore
+
+raw_outfile = Path(outdir) / "_prodigy_raw.txt"
+json_outfile = Path(outdir) / "_prodigy.json"
+tsv_outfile = Path(outdir) / "_prodigy.tsv"
+
+# log to the raw_outfile
+logging.basicConfig(level=logging.INFO, stream=sys.stdout, format="%(message)s")
+logger = logging.getLogger("Prodigy")
+
+if isinstance(selection, str):
+    selection = [selection]
+
+struct_path = check_path(infile)
+
+# parse structure
+structure, n_chains, n_res = parse_structure(struct_path)
+logger.info(
+    "[+] Parsed structure file {0} ({1} chains, {2} residues)".format(
+        structure.id, n_chains, n_res
+    )
+)
+prodigy = Prodigy(structure, selection, temperature)
+prodigy.predict(distance_cutoff=distance_cutoff, acc_threshold=acc_threshold)
+prodigy.print_prediction(outfile=raw_outfile, quiet=False)
+
+# Print out interaction network
+if contact_list:
+    prodigy.print_contacts(f"{outdir}/prodigy.ic")
+
+# Print out interaction network
+if pymol_selection:
+    prodigy.print_pymol_script(f"{outdir}/prodigy.pml")
+
+# [+] Reading structure file: <path/to/structure.cif>
+# [+] Parsed structure file <structure> (4 chains, 411 residues)
+# [+] No. of intermolecular contacts: 191
+# [+] No. of charged-charged contacts: 17
+# [+] No. of charged-polar contacts: 18
+# [+] No. of charged-apolar contacts: 60
+# [+] No. of polar-polar contacts: 5
+# [+] No. of apolar-polar contacts: 41
+# [+] No. of apolar-apolar contacts: 50
+# [+] Percentage of apolar NIS residues: 33.90
+# [+] Percentage of charged NIS residues: 30.48
+# [++] Predicted binding affinity (kcal.mol-1):    -21.3
+# [++] Predicted dissociation constant (M) at 25.0˚C:  2.3e-16
+
+output = {}
+with open(raw_outfile, "r") as f:
+    for line in f:
+        if line.startswith("[+"):
+            line = line.lstrip("[").lstrip("+").lstrip("]").lstrip()
+            if line.startswith("Reading structure file"):
+                continue
+            if line.startswith("Parsed structure file"):
+                continue
+
+            key, value = line.split(":", 1)
+            key = key.strip()
+            value = value.strip()
+            if key == "No. of intermolecular contacts":
+                output["nIC"] = int(value)
+            elif key == "No. of charged-charged contacts":
+                output["nCCC"] = int(value)
+            elif key == "No. of charged-polar contacts":
+                output["nCPC"] = int(value)
+            elif key == "No. of charged-apolar contacts":
+                output["nCAPC"] = int(value)
+            elif key == "No. of polar-polar contacts":
+                output["nPPC"] = int(value)
+            elif key == "No. of apolar-polar contacts":
+                output["nAPPC"] = int(value)
+            elif key == "No. of apolar-apolar contacts":
+                output["nAPAPC"] = int(value)
+            elif key.startswith("Percentage of apolar NIS residues"):
+                output["pANISR"] = float(value)
+            elif key.startswith("Percentage of charged NIS residues"):
+                output["pCNISR"] = float(value)
+            elif key.startswith("Predicted binding affinity"):
+                output["BindingAffinity"] = float(value)
+            elif key.startswith("Predicted dissociation constant"):
+                output["DissociationConstant"] = float(value)
+
+with open(json_outfile, "w") as f:
+    json.dump(output, f, indent=2)
+
+with open(tsv_outfile, "w") as f:
+    f.write("\t".join(output.keys()) + "\n")
+    f.write("\t".join(map(str, output.values())) + "\n")
+
+if outtype == "json":
+    json_outfile.rename(outfile)
+    json_outfile.symlink_to(outfile)
+elif outtype == "tsv":
+    tsv_outfile.rename(outfile)
+    tsv_outfile.symlink_to(outfile)
+else:
+    raw_outfile.rename(outfile)
+    raw_outfile.symlink_to(outfile)