[ERROR] failed to open file 'funannotate2-db/mito.refseq.fasta': No such file or directory

[marcinschmidt]

Hi,
I have installed funannotate2 in a singularity container and run the tutorial,
in funannotate2 predict ... I got an info:
Mitochondrial refseq database is not installed, unable to filter contigs
however, it looked like the database was installed sucessfully. So I forced to reinstall mito database
and got:
Downloading NCBI RefSeq mitochondrial genomes
[ERROR] failed to open file 'funannotate2-db/mito.refseq.fasta': No such file or directory

the retrieved for download link from GitHub Repo
"mito": "https://ftp.ncbi.nlm.nih.gov/refseq/release/mitochondrion/mitochondrion.1.1.genomic.fna.gz"
has an archive that is not converted into the mito.refseq.fasta file.

[marcinschmidt]

I have inspected the problem further and found that the mito database was downloaded successfully:

--2025-07-09 11:10:02--  https://ftp.ncbi.nlm.nih.gov/refseq/release/mitochondrion/mitochondrion.1.1.genomic.fna.gz
Resolving ftp.ncbi.nlm.nih.gov (ftp.ncbi.nlm.nih.gov)... 130.14.250.10, 130.14.250.12, 130.14.250.13, ...
Connecting to ftp.ncbi.nlm.nih.gov (ftp.ncbi.nlm.nih.gov)|130.14.250.10|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 180744545 (172M) [application/x-gzip]
Saving to: 'funannotate2-db/mito.refseq.fasta'
     0K .......... .......... .......... .......... ..........  0%  249K 11m47s
(...)
176500K ........                                              100%  332M=5.5s

2025-07-09 11:10:08 (31.5 MB/s) - 'funannotate2-db/mito.refseq.fasta' saved [180744545/180744545]

but I didn't find the file in the database directory nor the mito.mmi. However, the other file mito.refseq.version' was present.

Resolving ftp.ncbi.nlm.nih.gov (ftp.ncbi.nlm.nih.gov)... 130.14.250.7, 130.14.250.10, 130.14.250.12, ...
Connecting to ftp.ncbi.nlm.nih.gov (ftp.ncbi.nlm.nih.gov)|130.14.250.7|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 4
Saving to: 'funannotate2-db/mito.refseq.version'

     0K                                                       100% 10.4M=0s

2025-07-09 11:10:08 (10.4 MB/s) - 'funannotate2-db/mito.refseq.version' saved [4/4]

[ERROR] failed to open file 'funannotate2-db/mito.refseq.fasta': No such file or directory
[92m[Jul 09 11:10 AM] [0m[38;20mDatabase installation complete:
{
  "merops": {
    "type": "diamond",
    "db": "funannotate2-db/merops.dmnd",
    "version": "12.5",
    "date": "2023-01-19",
    "n_records": 5098,
    "md5": "1e75e1-5f29c85284c07"
  },
  "uniprot": {
    "type": "diamond",
    "db": "funannotate2-db/uniprot.dmnd",
    "version": "2025_03",
    "date": "2025-06-18",
    "n_records": 573661,
    "md5": "97-637df6d8d6873"
  },
  "dbCAN": {
    "type": "hmmer3",
    "db": "funannotate2-db/dbCAN.hmm",
    "version": "13.0",
    "date": "2024-08-14",
    "n_records": 826,
    "md5": "72cf7b3-61f9e3fb25183-gzip"
  },
  "pfam": {
    "type": "hmmer3",
    "db": "funannotate2-db/Pfam-A.hmm",
    "version": "37.4",
    "date": "2025-03",
    "n_records": 24736,
    "md5": "73-6310be2aa89ec"
  },
  "go": {
    "type": "text",
    "db": "funannotate2-db/go.obo",
    "version": "2025-06-01",
    "date": "2025-06-01",
    "n_records": 48101,
    "md5": "f7842e21ea50c6a4b59e452730a710a8"
  },
  "mibig": {
    "type": "diamond",
    "db": "funannotate2-db/mibig.dmnd",
    "version": "3.1",
    "date": "2025-07-09",
    "n_records": 41002,
    "md5": "6341db29-1b426e4"
  },
  "interpro": {
    "type": "xml",
    "db": "funannotate2-db/interpro.xml",
    "version": "106.0",
    "date": "2025-06-19",
    "n_records": 48679,
    "md5": "28bf90-637dbaae039fb"
  },
  "gene2product": {
    "type": "text",
    "db": "funannotate2-db/ncbi_cleaned_gene_products.txt",
    "version": "1.96",
    "date": "2025-07-09",
    "n_records": 34469,
    "md5": "W/\"5e548df4fbf9f05ce471baf70fc73c864145b93c4f4236bf5da2502ec6281ee0"
  },
  "mito": {
    "type": "minimap2",
    "db": "funannotate2-db/mito.mmi",
    "version": "230",
    "date": "2025-07-09",
    "n_records": 18207,
    "md5": "4-6347128a5bf2e"
  }
}[0m
[92m[Jul 09 11:10 AM] [0m[38;20mfunannotate2.install module finished: peak memory usage=2.23 GiB[0m

when --force update it went smooth, but again:

--2025-07-09 11:10:23--  https://ftp.ncbi.nlm.nih.gov/refseq/release/RELEASE_NUMBER
Resolving ftp.ncbi.nlm.nih.gov (ftp.ncbi.nlm.nih.gov)... 130.14.250.10, 130.14.250.12, 130.14.250.13, ...
Connecting to ftp.ncbi.nlm.nih.gov (ftp.ncbi.nlm.nih.gov)|130.14.250.10|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 4
Saving to: 'funannotate2-db/mito.refseq.version'

     0K                                                       100% 9.88M=0s

2025-07-09 11:10:24 (9.88 MB/s) - 'funannotate2-db/mito.refseq.version' saved [4/4]

[ERROR] failed to open file 'funannotate2-db/mito.refseq.fasta': No such file or directory```

[marcinschmidt]

I am working on HPC, and there might be an issue that script goes ahead but download is not finished yet. I have changed a part of install.py (during singularity-build) to:

def mitoDB(wget=False):
    import time
    from pathlib import Path

    mitoFA = os.path.join(env["FUNANNOTATE2_DB"], "mito.refseq.fasta")
    mitoMMI = os.path.join(env["FUNANNOTATE2_DB"], "mito.mmi")
    mitoREF = os.path.join(env["FUNANNOTATE2_DB"], "mito.refseq.version")

    logger.info("Downloading NCBI RefSeq mitochondrial genomes")

    for x in [mitoFA, mitoMMI, mitoREF]:
        if checkfile(x):
            os.remove(x)

    # get md5 remote to store so update works properly
    md5 = calcmd5remote(DBURL.get("mito-release"))

    # download fasta + version file
    download(DBURL.get("mito"), mitoFA, wget=wget)
    download(DBURL.get("mito-release"), mitoREF, wget=wget)

    # Wait until mitoFA actually appears
    mito_path = Path(mitoFA)
    for i in range(5):
        if mito_path.exists():
            break
        logger.info(f"Waiting for file {mitoFA} to appear...")
        time.sleep(1)
    else:
        logger.error(f"File {mitoFA} still not found after waiting.")
        sys.exit(1)

    with io.open(mitoREF, encoding="utf8", errors="ignore") as infile:
        version = infile.readline().strip()

    # now create mmi database
    cmd = ["minimap2", "-x", "asm20", "-d", mitoMMI, mitoFA]
    result = subprocess.run(cmd, cwd=os.path.join(env["FUNANNOTATE2_DB"]), capture_output=True, text=True)
    if result.returncode != 0:
        logger.error(f"Failed to create minimap2 index:\n{result.stderr}")
        sys.exit(1)

    num_records = countfasta(mitoFA)

    # we can delete the fasta as not needed
    if checkfile(mitoFA):
        os.remove(mitoFA)

    return {
        "type": "minimap2",
        "db": mitoMMI,
        "version": version,
        "date": today,
        "n_records": num_records,
        "md5": md5,
    }

and it solves the problem that the mito.refseq.fasta file is missing. However, still the mito.mmi wasn't created. So, I created it 'manually' adding to my annotation script (after database download):
minimap2 -x asm20 -d $FUNANNOTATE2_DB/mito.mmi $FUNANNOTATE2_DB/mito.refseq.fasta
So, the funannotate2 install reported:

[92m[Jul 09 10:23 PM] [0m[31;20mFailed to create minimap2 index:
[ERROR] failed to open file 'funannotate2-db/mito.refseq.fasta': No such file or directory

but finally , funannotate2 predict did its job.

[nextgenusfs]

Strange. Thanks for reporting and finding a solution. I'll look a little closer and see if we are missing an absolute path or something.