origenomi 0.1.1__tar.gz

origenomi-0.1.1/PKG-INFO

@@ -0,0 +1,47 @@
+Metadata-Version: 2.4
+Name: origenomi
+Version: 0.1.1
+Summary: Trim circular overlaps, find dnaA/oriC, rotate sequences, and report.
+Author: You
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+Requires-Dist: biopython>=1.83
+Requires-Dist: numpy>=1.26
+Requires-Dist: matplotlib>=3.7
+
+# Origami
+
+Command-line tool that:
+- Trims duplicated terminal overlaps in circular assemblies (half2 → DB(half1) via BLAST).
+- Runs **one BLAST per DB** (dnaA and oriC) on the concatenated FASTA.
+- Keeps **top 3 dnaA** and **top 7 oriC** per record (dedup by `sseqid`, ordered by evalue→identity→coverage→bitscore).
+- Pairs dnaA–oriC via midpoint proximity (1% then 5% of record length, else closest), computes AT/GC on the oriC segment, and chooses the pair with **highest AT/GC ratio** (tie → higher AT%).
+- Rotates so the **earlier** of dnaA/oriC starts the sequence (or the single site if only one found).
+- Writes exactly two outputs: `origami_<prefix>.fna` and `origami_<prefix>_report.txt`.
+
+## Install
+```bash
+pip install -e .
+```
+
+## Usage
+
+### Full pipeline
+```bash
+origami run -i genome.fna --dnaA-db /path/new_dnaA_DB --oric-db /path/DoriC_DB
+```
+
+### Trim only
+```bash
+origami trim -i genome.fna
+```
+
+### OriC only (no trim)
+```bash
+origami oric -i genome.fna --dnaA-db /path/new_dnaA_DB --oric-db /path/DoriC_DB
+```
+
+## Outputs
+1. origami_<prefix>.fna
+2. origami_<prefix>_report.txt
+3. Temps in ./temp/<prefix>/, cleaned unless --keep-temp.

origenomi-0.1.1/README.md

@@ -0,0 +1,36 @@
+# Origami
+
+Command-line tool that:
+- Trims duplicated terminal overlaps in circular assemblies (half2 → DB(half1) via BLAST).
+- Runs **one BLAST per DB** (dnaA and oriC) on the concatenated FASTA.
+- Keeps **top 3 dnaA** and **top 7 oriC** per record (dedup by `sseqid`, ordered by evalue→identity→coverage→bitscore).
+- Pairs dnaA–oriC via midpoint proximity (1% then 5% of record length, else closest), computes AT/GC on the oriC segment, and chooses the pair with **highest AT/GC ratio** (tie → higher AT%).
+- Rotates so the **earlier** of dnaA/oriC starts the sequence (or the single site if only one found).
+- Writes exactly two outputs: `origami_<prefix>.fna` and `origami_<prefix>_report.txt`.
+
+## Install
+```bash
+pip install -e .
+```
+
+## Usage
+
+### Full pipeline
+```bash
+origami run -i genome.fna --dnaA-db /path/new_dnaA_DB --oric-db /path/DoriC_DB
+```
+
+### Trim only
+```bash
+origami trim -i genome.fna
+```
+
+### OriC only (no trim)
+```bash
+origami oric -i genome.fna --dnaA-db /path/new_dnaA_DB --oric-db /path/DoriC_DB
+```
+
+## Outputs
+1. origami_<prefix>.fna
+2. origami_<prefix>_report.txt
+3. Temps in ./temp/<prefix>/, cleaned unless --keep-temp.

origenomi-0.1.1/pyproject.toml

@@ -0,0 +1,29 @@
+[build-system]
+requires = ["setuptools>=68", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "origenomi"
+version = "0.1.1"
+description = "Trim circular overlaps, find dnaA/oriC, rotate sequences, and report."
+authors = [{ name = "You" }]
+readme = "README.md"
+requires-python = ">=3.9"
+dependencies = [
+  "biopython>=1.83",
+  "numpy>=1.26",
+  "matplotlib>=3.7"
+  ]
+
+[project.scripts]
+origenomi = "origami.cli:main"
+
+[tool.setuptools]
+package-dir = {"" = "src"}
+packages = ["origami"]
+
+[tool.setuptools.package-data]
+origami = [
+  "database/**/"
+]
+

origenomi-0.1.1/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

origenomi-0.1.1/src/origami/blast.py

@@ -0,0 +1,122 @@
+import subprocess
+from subprocess import DEVNULL
+
+# === Annotation (dnaA / oriC) ===
+# Keep your exact outfmt for annotation BLASTs
+OUTFMT_ANN = (
+    "7 qseqid sseqid qstart qend sstart send qlen slen "
+    "pident evalue bitscore mismatch gaps length"
+)
+
+def run_blast_annotation(
+    query_fasta: str,
+    db_prefix: str,
+    out_path: str,
+    evalue: str = "1e-5",
+    quiet: bool = True,
+) -> None:
+    """
+    One BLAST per DB (dnaA, oriC) over the concatenated FASTA.
+    """
+    args = [
+        "blastn",
+        "-query", query_fasta,
+        "-db", db_prefix,
+        "-out", out_path,
+        "-outfmt", OUTFMT_ANN,
+        "-evalue", str(evalue),
+    ]
+    subprocess.run(
+        args,
+        check=True,
+        stdout=DEVNULL if quiet else None,
+        stderr=DEVNULL if quiet else None,
+    )
+
+def parse_blast_annotation(path: str):
+    """
+    Parse annotation outfmt into dict[qseqid] -> list[hit dicts].
+    """
+    hits = {}
+    with open(path) as f:
+        for line in f:
+            if not line.strip() or line.startswith("#"):
+                continue
+            (qid, sid, qstart, qend, sstart, send,
+             qlen, slen, pident, evalue, bitscore,
+             mismatch, gaps, length) = line.strip().split()
+
+            qstart, qend = int(qstart), int(qend)
+            sstart, send = int(sstart), int(send)
+
+            entry = {
+                "sseqid": sid,
+                "qstart": min(qstart, qend),
+                "qend":   max(qstart, qend),
+                "sstart": min(sstart, send),
+                "send":   max(sstart, send),
+                "qlen": int(qlen),
+                "slen": int(slen),
+                "length": int(length),
+                "pident": float(pident),
+                "evalue": float(evalue),
+                "bitscore": float(bitscore),
+                "mismatch": int(mismatch),
+                "gaps": int(gaps),
+            }
+            hits.setdefault(qid, []).append(entry)
+    return hits
+
+
+# === Trim (half2 → DB(half1)) ===
+# Minimal columns, strict flags; we sort by length and a couple tie-breakers
+OUTFMT_TRIM = "6 qseqid sseqid qstart qend sstart send qlen slen pident length evalue"
+
+def run_blast_trim(
+    query_fasta: str,
+    db_prefix: str,
+    out_path: str,
+    evalue: str = "1e-10",
+    quiet: bool = True,
+) -> None:
+    """
+    BLAST for terminal overlap detection (half2 vs DB(half1)).
+    """
+    args = [
+        "blastn",
+        "-query", query_fasta,
+        "-db", db_prefix,
+        "-out", out_path,
+        "-outfmt", OUTFMT_TRIM,
+        "-evalue", str(evalue),
+        "-dust", "no",
+        "-soft_masking", "false",
+        "-max_hsps", "1",
+    ]
+    subprocess.run(
+        args,
+        check=True,
+        stdout=DEVNULL if quiet else None,
+        stderr=DEVNULL if quiet else None,
+    )
+
+def parse_blast_trim(path: str):
+    """
+    Parse trim outfmt into a flat list of hits (dicts).
+    """
+    rows = []
+    with open(path) as f:
+        for line in f:
+            if not line.strip() or line.startswith("#"):
+                continue
+            (qid, sid, qstart, qend, sstart, send,
+             qlen, slen, pident, length, evalue) = line.strip().split()
+            rows.append({
+                "qseqid": qid, "sseqid": sid,
+                "qstart": int(qstart), "qend": int(qend),
+                "sstart": int(sstart), "send": int(send),
+                "qlen": int(qlen), "slen": int(slen),
+                "pident": float(pident), "length": int(length),
+                "evalue": float(evalue),
+            })
+    return rows

origenomi-0.1.1/src/origami/circos_tools.py

@@ -0,0 +1,272 @@
+import os
+import re
+import numpy as np
+import matplotlib.pyplot as plt
+from Bio import SeqIO
+
+#  SHARED FUNCTION
+def fast_gc_skew(seq, win=1000):
+    """Compute GC skew using cumulative NumPy operations."""
+    seq = np.frombuffer(seq.upper().encode(), dtype='S1')
+    g = (seq == b'G').astype(int)
+    c = (seq == b'C').astype(int)
+
+    cum_g = np.cumsum(g)
+    cum_c = np.cumsum(c)
+
+    skew = np.empty(len(seq) // win)
+
+    for i in range(len(skew)):
+        start = i * win
+        end = start + win
+
+        g_win = cum_g[end - 1] - (cum_g[start - 1] if start > 0 else 0)
+        c_win = cum_c[end - 1] - (cum_c[start - 1] if start > 0 else 0)
+
+        denom = g_win + c_win
+        skew[i] = (g_win - c_win) / denom if denom > 0 else 0
+
+    return skew
+
+
+def read_positions(report):
+    """Extract dnaA and oriC coordinates from Origami report."""
+    dnaA = None
+    oriC_list = []
+
+    with open(report) as f:
+        for line in f:
+            # dnaA coordinates
+            if "Chosen dnaA:" in line:
+                next(f)
+                coords_line = next(f).strip()
+                m = re.search(r"Coords:\s*(\d+)\s*-\s*(\d+)", coords_line)
+                if m:
+                    dnaA = int(m.group(1))
+
+            # oriC coordinates (can be multiple)
+            if "Chosen oriC:" in line:
+                next(f)
+                coords_line = next(f).strip()
+                m = re.findall(r"(\d+)\s*-\s*(\d+)", coords_line)
+                for a, b in m:
+                    oriC_list.append((int(a), int(b)))
+
+    return dnaA, oriC_list
+
+
+# ============================================================
+# 1) NON-ROTATED CIRCOS PLOT  →  plot_circos()
+# ============================================================
+
+def plot_circos(fna_path, report_path, out_dir):
+    """
+    This is the NON-rotated circos plot (Version B).
+    Used for original genome orientation.
+    """
+    record = next(SeqIO.parse(fna_path, "fasta"))
+    seq = str(record.seq)
+    genome_len = len(seq)
+    print(f"Genome Length in non rotated circos:{genome_len}")
+    dnaA, oriCs = read_positions(report_path)
+    oriC = oriCs[0][0] if oriCs else None
+
+    # Compute GC skew
+    skew = fast_gc_skew(seq, win=1000)
+    angles = np.linspace(0, 2*np.pi, len(skew), endpoint=False)
+
+    # Polar plot setup
+    fig, ax = plt.subplots(figsize=(10, 10), subplot_kw={'projection': 'polar'})
+    ax.set_theta_direction(-1)
+    ax.set_theta_offset(np.pi / 2)
+    ax.set_axis_off()
+
+    # Outer genome ring
+    outer_r = 1.0
+    ax.plot(np.linspace(0, 2*np.pi, 2000),
+            [outer_r] * 2000,
+            color='navy', lw=4)
+
+    # GC skew bars
+    bar_r = 0.8
+    bar_scale = 0.25
+
+    ax.bar(
+        angles[skew > 0],
+        skew[skew > 0] * bar_scale,
+        width=2*np.pi/len(skew),
+        bottom=bar_r,
+        color='lightblue',
+        linewidth=0
+    )
+    ax.bar(
+        angles[skew < 0],
+        -skew[skew < 0] * bar_scale,
+        width=2*np.pi/len(skew),
+        bottom=bar_r - bar_scale * np.abs(skew[skew < 0]),
+        color='orange',
+        linewidth=0
+    )
+
+    # Ticks (always 28)
+    num_ticks = 28
+    tick_spacing = genome_len / num_ticks
+    label_offset = 0.10
+
+    for i in range(num_ticks):
+        pos = tick_spacing * i
+        theta = 2 * np.pi * (pos / genome_len)
+
+        ax.plot([theta, theta], [outer_r, outer_r + 0.03],
+                color='navy', lw=1.5)
+        ax.text(theta, outer_r + label_offset,
+                f"{pos/1e6:.2f} Mb",
+                ha='center', va='center', fontsize=9)
+
+    # dnaA / oriC markers
+    if dnaA:
+        theta_dnaA = 2 * np.pi * (dnaA / genome_len)
+        ax.plot([theta_dnaA], [outer_r], marker='s',
+                color='blue', markersize=10)
+
+    if oriC:
+        theta_oriC = 2 * np.pi * (oriC / genome_len)
+        ax.plot([theta_oriC], [outer_r], marker='o',
+                color='red', markersize=10)
+
+    # Legend
+    fig.text(0.05, 0.93, u'\u25A0 dnaA', color='blue', fontsize=13, fontweight='bold')
+    fig.text(0.05, 0.89, u'\u25CF oriC', color='red', fontsize=13, fontweight='bold')
+
+    # Title text
+    desc = record.description.split(",")[0]
+    parts = desc.split(" ", 1)
+    line1 = parts[0]
+    line2 = parts[1] if len(parts) > 1 else ""
+
+    ax.text(0, 0,
+            f"{line1}\n{line2}\n({genome_len:,} bp)",
+            ha='center', va='center',
+            fontsize=10, fontweight='bold')
+
+    # Save
+    base = os.path.splitext(os.path.basename(fna_path))[0]
+    png = os.path.join(out_dir, f"{base}_final_circos_v9.png")
+    svg = os.path.join(out_dir, f"{base}_final_circos_v9.svg")
+
+    plt.savefig(png, dpi=400, bbox_inches='tight')
+    plt.savefig(svg, bbox_inches='tight')
+    plt.close()
+
+    return png  # main.py expects a PNG path
+
+
+# ============================================================
+# 2) ROTATED CIRCOS PLOT  →  plot_rotated_circos()
+# ============================================================
+
+def plot_rotated_circos(fna_path, report_path, out_dir):
+    """
+    ROTATED circos plot (Version A).
+    Adjusts genome so the origin (dnaA or oriC) starts at 0°.
+    """
+    record = next(SeqIO.parse(fna_path, "fasta"))
+    seq = str(record.seq)
+    genome_len = len(seq)
+    print(f"Genome Length in rotated circos:{genome_len}")
+    dnaA, oriC_list = read_positions(report_path)
+    oriC = oriC_list[0][0] if oriC_list else None
+
+    # Determine shift point: whichever (dnaA or oriC) is further.
+    a, b = sorted([dnaA, oriC])
+    d = b - a
+
+    if d > genome_len / 2:
+        shift_bp = b
+        rotated_from = "oriC" if b == oriC else "dnaA"
+    else:
+        shift_bp = a
+        rotated_from = "oriC" if a == oriC else "dnaA"
+
+    # Compute GC skew and rotate
+    skew = fast_gc_skew(seq, win=1000)
+    bins = len(skew)
+    bin_size = genome_len / bins
+    # shift_bins = int(shift_bp / bin_size)
+    # skew = np.roll(skew, -shift_bins)
+
+    # Polar plot
+    fig, ax = plt.subplots(figsize=(10, 10), subplot_kw={'projection': 'polar'})
+    ax.set_theta_direction(-1)
+    ax.set_theta_offset(np.pi / 2)
+    ax.set_axis_off()
+
+    # Outer ring
+    outer_r = 1.0
+    ax.plot(np.linspace(0, 2*np.pi, 2000),
+            [outer_r] * 2000,
+            color='navy', lw=4)
+
+    # GC skew bars
+    r_inner = 0.8
+    bar_scale = 0.25
+    angles = np.linspace(0, 2*np.pi, bins, endpoint=False)
+
+    ax.bar(angles[skew > 0], skew[skew > 0] * bar_scale,
+           width=2*np.pi/bins, bottom=r_inner,
+           color='lightblue', linewidth=0)
+
+    ax.bar(angles[skew < 0], -skew[skew < 0] * bar_scale,
+           width=2*np.pi/bins,
+           bottom=r_inner - bar_scale*np.abs(skew[skew < 0]),
+           color='orange', linewidth=0)
+
+    # Ticks (28)
+    num_ticks = 28
+    tick_spacing = genome_len / num_ticks
+    label_offset = 0.08
+
+    for i in range(num_ticks):
+        pos = tick_spacing * i
+        theta = 2 * np.pi * i / num_ticks
+        ax.plot([theta, theta], [outer_r, outer_r + 0.03],
+                color='navy', lw=1.5)
+        ax.text(theta, outer_r + label_offset,
+                f"{pos/1e6:.2f} Mb",
+                ha='center', va='center', fontsize=9)
+
+    # Rotated marker positions
+    dnaA_rot = (dnaA - shift_bp) % genome_len
+    oriC_rot = (oriC - shift_bp) % genome_len
+
+    ax.plot([2*np.pi*(dnaA_rot/genome_len)], [outer_r],
+            marker='s', color='blue', markersize=10)
+    ax.plot([2*np.pi*(oriC_rot/genome_len)], [outer_r],
+            marker='o', color='red', markersize=10)
+
+    # Center label
+    desc = record.description.split(",")[0]
+    parts = desc.split(" ", 1)
+    line1 = parts[0]
+    line2 = parts[1] if len(parts) > 1 else ""
+
+    ax.text(0, 0,
+            f"{line1}\n{line2}\n({genome_len:,} bp)",
+            ha='center', va='center',
+            fontsize=10, fontweight='bold')
+
+    # Legend
+    fig.text(0.05, 0.93, u'\u25A0 dnaA', color='blue', fontsize=13, fontweight='bold')
+    fig.text(0.05, 0.89, u'\u25CF oriC', color='red', fontsize=13, fontweight='bold')
+
+    # Save
+    base = os.path.splitext(os.path.basename(fna_path))[0]
+    png = os.path.join(out_dir, f"{base}_rotated_circular_gcskew.png")
+    svg = os.path.join(out_dir, f"{base}_rotated_circular_gcskew.svg")
+
+    plt.savefig(png, dpi=300, bbox_inches='tight')
+    plt.savefig(svg, bbox_inches='tight')
+    plt.close()
+
+    # Return the PNG, the chosen marker, and the coordinate
+    return png, rotated_from, shift_bp

origenomi-0.1.1/src/origami/circularize.py

@@ -0,0 +1,33 @@
+
+def prepare_chromosome_sequence(header: str, sequence: str, duplicate_bp: int = 4000) -> str | None:
+    """
+    Prepares a bacterial chromosome sequence for BLAST:
+    - Skips plasmids
+    - Adds the first `duplicate_bp` bases at the end to handle circular overlap
+    """
+    # checking for plasmid in the header
+    if "plasmid" in header.lower():
+        return sequence  # skip plasmids
+
+    # Make sure we don't exceed sequence length
+    bp_to_duplicate = min(len(sequence), duplicate_bp)
+    extended_seq = sequence + sequence[:bp_to_duplicate]
+    return extended_seq
+
+
+def trim_extended_if_needed(seq: str, hit: dict, extension_len: int = 4000):
+    """
+    Adjust sequence and coordinates:
+      - If hit is in the extension (last extension_len bp), 
+        drop first extension_len bp.
+      - Otherwise, drop artificial tail.
+    """
+    seq_len = len(seq)
+    orig_len = seq_len - extension_len
+
+    if hit["qend"] > orig_len:
+        # Hit was in extension → drop front
+        return seq[extension_len:], "front"
+    else:
+        # No hit in extension → drop tail
+        return seq[:-extension_len], "tail"

origenomi-0.1.1/src/origami/cli.py

@@ -0,0 +1,61 @@
+import argparse
+import os
+from origami.workflows import run_full, run_trim_only, run_oric_only
+from origami.logging_setup import setup_logging
+from importlib.resources import files
+
+# Default DB paths (relative to project root)
+DNAA_PREFIX = "dnaA_clustered100_DB"
+ORIC_PREFIX = "OriC_clustered100_DB"
+
+def get_db_prefixes():
+    db_root = files("origami.data.clustered_DB")
+    dnaA_prefix = db_root / "dnaA_clustered100_DB"
+    oric_prefix = db_root / "OriC_clustered100_DB"
+    return str(dnaA_prefix), str(oric_prefix)
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        prog="origenomi",
+        description="Trim circular overlaps, find dnaA/oriC, rotate, and report."
+    )
+    subparsers = parser.add_subparsers(dest="command")
+
+    def add_common(p):
+        p.add_argument("-i", "--input", required=True, help="Input FASTA (single or multi-FASTA)")
+        p.add_argument("-o","--out-dir", default=".", help="Output directory (default: .)")
+        p.add_argument("--keep-temp", action="store_true", help="Keep temp files")
+        p.add_argument("--verbose", action="store_true", help="Verbose logging")
+        p.add_argument("--keep-plot", action="store_true", help="Keep the plots")
+
+    # Full pipeline
+    p_run = subparsers.add_parser("run", help="Trim + dnaA/oriC + rotate + report")
+    add_common(p_run)
+
+    # Trim-only
+    p_trim = subparsers.add_parser("trim", help="Trim duplicated terminal overlaps only")
+    add_common(p_trim)
+
+    # OriC-only
+    p_oric = subparsers.add_parser("oric", help="dnaA/oriC + rotate on original input")
+    add_common(p_oric)
+
+    args = parser.parse_args()
+
+    # Inject default DB paths automatically
+    args.dnaA_db, args.oric_db = get_db_prefixes()
+
+    args.out_dir = os.path.abspath(args.out_dir)
+    os.makedirs(args.out_dir, exist_ok=True)
+
+    logger = setup_logging(args.verbose)
+
+    if args.command == "run":
+        run_full(args, logger)
+    elif args.command == "trim":
+        run_trim_only(args, logger)
+    elif args.command == "oric":
+        run_oric_only(args, logger)
+    else:
+        parser.print_help()

origenomi-0.1.1/src/origami/config.py

@@ -0,0 +1,5 @@
+DNA_TOP_K = 3
+ORIC_TOP_K = 7
+TIER1_THRESHOLD = 0.01  # 1%
+TIER2_THRESHOLD = 0.05  # 5%
+WRAP_WIDTH = 60