Skip to content

augustus

Augustus gene prediction integration.

augustus

augustus_functional()

Run augustus with --proteinprofile on a small bundled example and check whether the binary is usable on this host. Returns True if augustus produced the expected gene prediction marker, False otherwise.

On failure, a diagnostic line is written to stderr distinguishing: * augustus binary missing from PATH * augustus killed by a signal (typically SIGILL — the binary contains CPU instructions this host cannot run, e.g. a linux/amd64 bioconda build under Apple Silicon / Rosetta 2, or pre-AVX hardware) * augustus exited non-zero with its own "augustus: ERROR" message * augustus exited cleanly but produced no recognizable output

Source code in buscolite/augustus.py 
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
def augustus_functional():
    """
    Run augustus with --proteinprofile on a small bundled example and check
    whether the binary is usable on this host. Returns True if augustus
    produced the expected gene prediction marker, False otherwise.

    On failure, a diagnostic line is written to stderr distinguishing:
      * augustus binary missing from PATH
      * augustus killed by a signal (typically SIGILL — the binary contains
        CPU instructions this host cannot run, e.g. a linux/amd64 bioconda
        build under Apple Silicon / Rosetta 2, or pre-AVX hardware)
      * augustus exited non-zero with its own "augustus: ERROR" message
      * augustus exited cleanly but produced no recognizable output
    """
    datadir = os.path.join(os.path.dirname(__file__), "data")
    cmd = [
        "augustus",
        "--species=human",
        "--softmasking=1",
        "--gff3=on",
        "--UTR=off",
        "--proteinprofile={}".format(os.path.join(datadir, "HsDHC.prfl")),
        "--stopCodonExcludedFromCDS=False",
        os.path.join(datadir, "example.fa"),
    ]
    try:
        proc = subprocess.run(
            cmd,
            stdout=subprocess.PIPE,
            stderr=subprocess.PIPE,
            universal_newlines=True,
            timeout=120,
        )
    except FileNotFoundError:
        sys.stderr.write("augustus binary not found on PATH\n")
        return False
    except subprocess.TimeoutExpired:
        sys.stderr.write("augustus --proteinprofile self-test timed out after 120s\n")
        return False

    stdout = (proc.stdout or "").strip()
    stderr = (proc.stderr or "").strip()
    rc = proc.returncode

    if rc is not None and rc < 0:
        sig = -rc
        try:
            sig_name = signal.Signals(sig).name
        except (ValueError, AttributeError):
            sig_name = ""
        sys.stderr.write(
            "augustus --proteinprofile self-test was killed by signal {} ({}). "
            "This typically means the augustus binary contains CPU "
            "instructions your host does not support — most commonly seen "
            "when running a linux/amd64 docker image on Apple Silicon via "
            "Rosetta 2, or on older x86_64 hardware lacking AVX/AVX2/BMI2. "
            "Run on real x86_64 hardware or use a native install.\n".format(sig, sig_name)
        )
        return False

    if stderr.startswith("augustus: ERROR"):
        sys.stderr.write(stderr + "\n")
        return False

    for line in stdout.split("\n"):
        if line.strip().startswith("# start gene g1"):
            return True

    sys.stderr.write(
        "augustus --proteinprofile self-test exited rc={} but produced no "
        "gene prediction. stderr was: {!r}\n".format(rc, stderr[:500])
    )
    return False