cutseq 0.0.1__py3-none-any.whl

cutseq/run.py

@@ -0,0 +1,399 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+#
+# Copyright © 2024 Ye Chang yech1990@gmail.com
+# Distributed under terms of the GNU license.
+#
+# Created: 2024-04-19 18:57
+
+import argparse
+import logging
+import re
+import subprocess
+import sys
+
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s -  %(levelname)s - %(message)s",
+)
+
+
+def reverse_complement(b):
+    return "".join(
+        [dict(zip("ATGCNatgcn", "TACGNtacgn"))[x] for x in b[::-1] if x in "ATGCNatgcn"]
+    )
+
+
+def remove_fq_suffix(f):
+    suffixes = [
+        "_R1_001.fastq",
+        "_R2_001.fastq",
+        "_R1.fastq",
+        "_R2.fastq",
+        ".fastq",
+        ".fq",
+    ]
+    suffixes = [s + ".gz" for s in suffixes] + suffixes
+    for suffix in suffixes:
+        if f.endswith(suffix):
+            return f.removesuffix(suffix)
+    return f
+
+
+class BarcodeConfig:
+    def __init__(self, adapter=None):
+        self.strand = None
+        self.p5_fw = ""
+        self.p5_rc = ""
+        self.p7_fw = ""
+        self.p7_rc = ""
+        self.inline5_fw = ""
+        self.inline5_rc = ""
+        self.inline5 = 0
+        self.inline3_fw = ""
+        self.inline3_rc = ""
+        self.inline3 = 0
+        self.umi5 = 0
+        self.umi3 = 0
+        self.mask5 = 0
+        self.mask3 = 0
+        if adapter is not None:
+            self._parse_barcode(adapter)
+
+    def _parse_barcode(self, b):
+        m = re.match(
+            r"(?P<p5>[ATGCatgc]+)(\((?P<inline5>[ATGCatgc]+)\))?(?P<umi5>N*)(?P<mask5>X*)(?P<strand>-|>|<)(?P<mask3>X*)(?P<umi3>N*)(\((?P<inline3>[ATGCatgc]+)\))?(?P<p7>[ATGCatgc]+)",
+            b,
+        )
+        if m is None:
+            logging.error(f"barcode {b} is not valid")
+            sys.exit(1)
+        d = m.groupdict()
+        if d["inline5"] is None:
+            d["inline5"] = ""
+        if d["inline3"] is None:
+            d["inline3"] = ""
+        self.strand = "+" if d["strand"] == ">" else "-" if d["strand"] == "<" else None
+        self.p5_fw = d["p5"]
+        self.p5_rc = reverse_complement(d["p5"])
+        self.p7_fw = d["p7"]
+        self.p7_rc = reverse_complement(d["p7"])
+        self.inline5_fw = d["inline5"] if d["inline5"] else ""
+        self.inline5_rc = reverse_complement(d["inline5"]) if d["inline5"] else ""
+        self.inline5 = len(d["inline5"])
+        self.inline3_fw = d["inline3"] if d["inline3"] else ""
+        self.inline3_rc = reverse_complement(d["inline3"]) if d["inline3"] else ""
+        self.inline3 = len(d["inline3"])
+        self.umi5 = len(d["umi5"])
+        self.umi3 = len(d["umi3"])
+        self.mask5 = len(d["mask5"])
+        self.mask3 = len(d["mask3"])
+
+
+class CutadaptConfig:
+    def __init__(self):
+        self.rname_suffix = False
+        self.discarded_untrimmed = False
+        self.trim_polyA = False
+        self.min_length = 20
+        self.min_quality = 20
+        self.dry_run = False
+        self.threads = 1
+
+
+def run_cutadapt_PE(
+    input1, input2, output1, output2, discard1, discard2, barcode, settings
+):
+    cutadapt = f"cutadapt -j {settings.threads}"
+    steps = []
+    # step 1: remove adapter on the 5' end, artifact of template switching
+    if settings.rname_suffix:
+        config_rname = " --strip-suffix '/1' --strip-suffix '/2' --strip-suffix '.1' --strip-suffix '.2'"
+    else:
+        config_rname = ""
+    steps.append(
+        f"{cutadapt}{config_rname} -e 0.25 -n 2 -O 10 -g '{barcode.p5_fw};rightmost' -G '{barcode.p7_rc};rightmost' --interleaved {input1} {input2}"
+    )
+    # step 2: remove adapter on the 3' end, read though in the sequencing
+    steps.append(
+        f"{cutadapt} -e 0.2 -n 2 -O 3 -a '{barcode.p7_fw}' -A '{barcode.p5_rc}' --interleaved -"
+    )
+    # step 3: trim inline barcode
+    config_inline_args = []
+    if barcode.inline5 > 0:
+        config_inline_args.append(f"-g ^{barcode.inline5_fw} -U -{barcode.inline5}")
+    if barcode.inline3 > 0:
+        config_inline_args.append(f"-G ^{barcode.inline3_rc} -u -{barcode.inline3}")
+    if barcode.inline5 + barcode.inline3 > 0:
+        if settings.discarded_untrimmed:
+            config_inline_args.append(
+                f"--untrimmed-output={discard1} --untrimmed-paired-output={discard2}"
+            )
+        config_inline = " ".join(config_inline_args)
+        steps.append(f"{cutadapt} {config_inline} --interleaved -")
+    # step 4: extract UMI
+    if barcode.umi5 + barcode.umi3 > 0:
+        steps.append(
+            f"{cutadapt} -u {barcode.umi5} -u -{barcode.umi3} -U {barcode.umi3} -U -{barcode.umi5} --rename='{{id}}_{{r1.cut_prefix}}{{r2.cut_prefix}}' --interleaved -"
+        )
+    else:
+        steps.append(f"{cutadapt} --rename='{{id}}' --interleaved -")
+    # step 5: mask tail in the RNA, which might be artifact of RT
+    if barcode.mask5 + barcode.mask3 > 0:
+        steps.append(
+            f"{cutadapt} -u {barcode.mask5} -u -{barcode.mask3} -U {barcode.mask3} -U -{barcode.mask5} --interleaved -"
+        )
+    # step 6: trim polyA
+    if settings.trim_polyA:
+        if barcode.strand == "+":
+            steps.append(
+                f"{cutadapt} -O 6 -e 0.15 -a 'A{{100}}' -G 'T{{100}}' --interleaved -"
+            )
+        elif barcode.strand == "-":
+            steps.append(
+                f"{cutadapt} -O 6 -e 0.15 -g 'T{{100}}' -A 'A{{100}}' --interleaved -"
+            )
+        else:
+            logging.info("No strand information provided, skip polyA trimming.")
+    # step 7: quality control, remove short reads
+    steps.append(
+        f"{cutadapt} -q {settings.min_quality} --max-n=0 -m {settings.min_length} --too-short-output={discard1} --too-short-paired-output={discard2} -o {output1} -p {output2} --interleaved -"
+    )
+
+    if settings.dry_run:
+        print(
+            " |\\\n".join([("    " + s if i > 0 else s) for i, s in enumerate(steps)])
+        )
+        process = subprocess.run("true", shell=True, capture_output=True)
+    else:
+        cmd = " | ".join(steps)
+        process = subprocess.run(cmd, shell=True, capture_output=True)
+    return process.stdout.decode(), process.stderr.decode()
+
+
+def run_cutadapt_SE(input1, output1, discard1, barcode, settings):
+    cutadapt = f"cutadapt -j {settings.threads}"
+    steps = []
+    # step 1: remove adapter on the 5' end, artifact of template switching
+    if settings.rname_suffix:
+        config_rname = " --strip-suffix '/1' --strip-suffix '.1'"
+    else:
+        config_rname = ""
+    steps.append(
+        f"{cutadapt}{config_rname} -e 0.25 -n 2 -O 10 -g '{barcode.p5_fw};rightmost' {input1}"
+    )
+    # step 2: remove adapter on the 3' end, read though in the sequencing
+    steps.append(f"{cutadapt} -e 0.2 -n 2 -O 3 -a '{barcode.p7_fw}' -")
+    # step 3: trim inline barcode
+    config_inline_args = []
+    if barcode.inline3 > 0:
+        config_inline_args.append(f"-a {barcode.inline3_fw}$")
+    if barcode.inline5 > 0:
+        config_inline_args.append(f"-g ^{barcode.inline5_fw}")
+    if barcode.inline5 + barcode.inline3 > 0:
+        if settings.discarded_untrimmed:
+            config_inline_args.append(f"--untrimmed-output={discard1}")
+
+        config_inline = " ".join(config_inline_args)
+        steps.append(f"{cutadapt} {config_inline} -")
+    # step 4: extract UMI
+    if barcode.umi5 + barcode.umi3 > 0:
+        steps.append(
+            f"{cutadapt} -u {barcode.umi5} -u -{barcode.umi3} --rename='{{id}}_{{cut_prefix}}{{cut_suffix}}' -"
+        )
+    # step 5: mask tail in the RNA, which might be artifact of RT
+    steps.append(f"{cutadapt} -u {barcode.mask5} -u -{barcode.mask3} -")
+    # step 6: trim polyA
+    if settings.trim_polyA:
+        if barcode.strand == "+":
+            steps.append(f"{cutadapt} -O 6 -e 0.15 -a 'A{{100}}' -")
+        elif barcode.strand == "-":
+            steps.append(f"{cutadapt} -O 6 -e 0.15 -g 'T{{100}}' -")
+        else:
+            logging.info("No strand information provided, skip polyA trimming.")
+    # step 7: quality control, remove short reads
+    steps.append(
+        f"{cutadapt} -q {settings.min_quality} --max-n=0 -m {settings.min_length} --too-short-output={discard1} -o {output1} -"
+    )
+
+    if settings.dry_run:
+        print(
+            " |\\\n".join([("    " + s if i > 0 else s) for i, s in enumerate(steps)])
+        )
+        process = subprocess.run("true", shell=True, capture_output=True)
+    else:
+        cmd = " | ".join(steps)
+        process = subprocess.run(cmd, shell=True, capture_output=True)
+    return process.stdout.decode(), process.stderr.decode()
+
+
+def run_cutseq(args):
+    barcode_config = BarcodeConfig(args.adapter)
+    settings = CutadaptConfig()
+    if args.with_rname_suffix:
+        settings.rname_suffix = True
+    if args.discarded_untrimmed:
+        settings.discarded_untrimmed = True
+    if args.trim_polyA:
+        settings.trim_polyA = True
+    settings.threads = args.threads
+    settings.min_length = args.min_length
+    settings.dry_run = args.dry_run
+    # Example command setup, you'll need to expand this based on your actual requirements
+    if len(args.input_file) == 1:
+        stdout, stderr = run_cutadapt_SE(
+            args.input_file[0],
+            args.output_file[0],
+            args.discard_file[0],
+            barcode_config,
+            settings,
+        )
+    else:
+        stdout, stderr = run_cutadapt_PE(
+            args.input_file[0],
+            args.input_file[1],
+            args.output_file[0],
+            args.output_file[1],
+            args.discard_file[0],
+            args.discard_file[1],
+            barcode_config,
+            settings,
+        )
+    print(stdout)
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Trim sequencing adapters from NGS data automatically."
+    )
+    # input file can be one or two for single or paired-end reads, but can not be more than two
+    parser.add_argument(
+        "input_file",
+        type=str,
+        nargs="+",
+        help="Input file path for NGS data, one or two files.",
+    )
+    # output file can be number of files matching the input files, if not provided it will generate based on the output suffix,
+    # if no output suffix provided it will generate based on the input file name
+    parser.add_argument(
+        "-a",
+        "--adapter",
+        type=str,
+        required=True,
+        help="Adapter sequence configuration.",
+    )
+    parser.add_argument(
+        "-O",
+        "--output-suffix",
+        type=str,
+        help="Output file suffix for keep trimmed data.",
+    )
+    parser.add_argument(
+        "-o",
+        "--output-file",
+        type=str,
+        nargs="+",
+        help="Output file path for keep trimmed data.",
+    )
+
+    # discard short reads
+    parser.add_argument(
+        "-D",
+        "--discard-file",
+        type=str,
+        nargs="+",
+        help="Output file path for discarded trimmed data.",
+    )
+    parser.add_argument(
+        "-m",
+        "--min-length",
+        type=int,
+        default=20,
+        help="Minimum length of the reads to keep.",
+    )
+    parser.add_argument(
+        "-q",
+        "--min-quality",
+        type=int,
+        default=20,
+        help="Minimum quality of the read tails in the reads to keep..",
+    )
+
+    parser.add_argument(
+        "--with-rname-suffix",
+        action="store_true",
+        help="R1 and R2 suffix cotains suffix. MGI platform.",
+    )
+    parser.add_argument(
+        "--discarded-untrimmed",
+        action="store_true",
+        help="Discard untrimmed reads (without inline barcode matching).",
+    )
+    parser.add_argument("--trim-polyA", action="store_true", help="Trim polyA tail.")
+
+    parser.add_argument(
+        "-t",
+        "--threads",
+        type=int,
+        default=1,
+        help="Number of threads to use for trimming.",
+    )
+    parser.add_argument(
+        "-n",
+        "--dry-run",
+        action="store_true",
+        help="Print command instead of running it.",
+    )
+    args = parser.parse_args()
+
+    if len(args.input_file) > 2:
+        raise ValueError("Input file can not be more than two.")
+
+    if args.output_file:
+        if len(args.output_file) != len(args.input_file):
+            raise ValueError("Output file should be same as input file.")
+    elif args.output_suffix:
+        if len(args.input_file) == 1:
+            args.output_file = [args.output_suffix + "_trimmed_R1.fastq.gz"]
+        else:
+            args.output_file = [
+                args.output_suffix + "_trimmed_R1.fastq.gz",
+                args.output_suffix + "_trimmed_R2.fastq.gz",
+            ]
+    else:
+        if len(args.input_file) == 1:
+            args.output_file = [
+                remove_fq_suffix(args.input_file[0]) + "_trimmed_R1.fastq.gz",
+            ]
+        else:
+            args.output_file = [
+                remove_fq_suffix(args.input_file[0]) + "_trimmed_R1.fastq.gz",
+                remove_fq_suffix(args.input_file[1]) + "_trimmed_R2.fastq.gz",
+            ]
+
+    if args.discard_file:
+        if len(args.discard_file) != len(args.input_file):
+            raise ValueError("Discard file should be same as input file.")
+    elif args.output_suffix:
+        if len(args.input_file) == 1:
+            args.discard_file = [args.output_suffix + "_discarded_R1.fastq.gz"]
+        else:
+            args.discard_file = [
+                args.output_suffix + "_discarded_R1.fastq.gz",
+                args.output_suffix + "_discarded_R2.fastq.gz",
+            ]
+    else:
+        if len(args.input_file) == 1:
+            args.discard_file = [
+                remove_fq_suffix(args.input_file[0]) + "_discarded_R1.fastq.gz",
+            ]
+        else:
+            args.discard_file = [
+                remove_fq_suffix(args.input_file[0]) + "_discarded_R1.fastq.gz",
+                remove_fq_suffix(args.input_file[1]) + "_discarded_R2.fastq.gz",
+            ]
+    run_cutseq(args)
+
+
+if __name__ == "__main__":
+    main()

cutseq-0.0.1.dist-info/METADATA

@@ -0,0 +1,23 @@
+Metadata-Version: 2.1
+Name: cutseq
+Version: 0.0.1
+Summary: Automatic cutadapter and barcode process for NGS data
+Home-page: https://github.com/y9c/cutseq
+License: MIT
+Keywords: bioinformatics,NGS,adapter,barcode,UMI
+Author: Ye Chang
+Author-email: yech1990@gmail.com
+Requires-Python: >=3.8,<4.0
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Requires-Dist: cutadapt (>=4.8,<5.0)
+Project-URL: Repository, https://github.com/y9c/cutseq
+Description-Content-Type: text/markdown
+
+# CutSeq
+

cutseq-0.0.1.dist-info/RECORD

@@ -0,0 +1,5 @@
+cutseq/run.py,sha256=WWQJKxd_tyc8Y4uQsjAo1fvHXd37skCLllqUEWJ-4p0,13807
+cutseq-0.0.1.dist-info/METADATA,sha256=JdXExUzHOvGGdNWj0COyVqgVY8HFJ1U6_QnzvVnTDb4,789
+cutseq-0.0.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+cutseq-0.0.1.dist-info/entry_points.txt,sha256=cz5WcGOzPAtTCE9mFfyDna3ofZv0NOeGtqE50rROBFU,42
+cutseq-0.0.1.dist-info/RECORD,,

cutseq-0.0.1.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: poetry-core 1.9.0
+Root-Is-Purelib: true
+Tag: py3-none-any

cutseq-0.0.1.dist-info/entry_points.txt

@@ -0,0 +1,3 @@
+[console_scripts]
+cutseq=cutseq.run:main
+