Outputs

A run writes its main results directly into --outdir, plus a JSON run log and (unless --removetmp) an archive of intermediates.

Main outputs

File	Contents
PREFIX.EEs.fa	Endogenous element nucleotide sequences.
PREFIX.EEs.tax.tsv	Endogenous element taxonomy table (one row per element).
PREFIX.EEs.flanks.fa	Endogenous elements plus flanking regions (--flank nt each side).
eefinder.log	JSON run summary (see below).

With --clean_masked, the mask-cleaned equivalents are written alongside them: PREFIX.EEs.cleaned.fa and PREFIX.EEs.cleaned.tax.tsv.

outdir/
├── eefinder.log
├── PREFIX.EEs.fa
├── PREFIX.EEs.tax.tsv
├── PREFIX.EEs.flanks.fa
├── PREFIX.EEs.cleaned.fa          # only with --clean_masked
├── PREFIX.EEs.cleaned.tax.tsv     # only with --clean_masked
└── tmp_files/                     # unless --removetmp

The taxonomy table (PREFIX.EEs.tax.tsv)

Column	Meaning
Element-ID	CONTIG:START-END identifier of the element (BED-style, 0-based half-open). The run prefix is not part of it — see the note below.
Sense	Strand of the element, pos or neg.
Protein-IDs	Accession(s) of the supporting reference protein(s), each followed by | and the percent identity of its hit.
Protein-Products	Product name(s) from the metadata CSV.
Molecule_type	Genome composition of the source (e.g. ssRNA(+), dsDNA-RT).
Family / Genus / Species	Taxonomic assignment from the metadata CSV.
Host	Host recorded for the reference.
Overlaped_Element_ID	Comma-separated element(s) of a different family this one overlaps (within 100 nt), if any.
tag	overlaped when Overlaped_Element_ID is non-empty, otherwise unique.
Average_pident	Mean of the percent identities listed in Protein-IDs, rounded to one decimal.

An element merged from several fragments carries all its supporting references: Protein-IDs entries are separated by |, while the parallel taxonomy columns (Protein-Products, Genus, Species, Host) join their values with AND.

Element-ID           Sense  Protein-IDs                                Family          Genus                            tag        Average_pident
ctg_1913:1754-2689   neg    YP_006732334.1|30.573 | NP_068729.1|28.239 Caulimoviridae  Soymovirus AND Caulimovirus      overlaped  29.4
ctg_1913:70282-70728 neg    YP_009666257.1|30.719                      Chuviridae      Scarabeuvirus                    unique     30.7

Missing metadata is filled rather than left blank: an unresolved Family, Genus or Species becomes Unclassified, and an unresolved Host becomes Undefined.

Note

The two places an element is named use two spellings. The taxonomy table’s Element-ID drops the run prefix (ctg_1913:1754-2689), while the EEs.fa / EEs.flanks.fa headers carry it (Ae_aeg_Aag2_ctg_1913/ctg_1913:1754-2689), so the sequences cross-reference the genome the run was given. Prepend {prefix}/ to an Element-ID to match a FASTA record.

The cleaned outputs (--clean_masked)

PREFIX.EEs.cleaned.fa holds the subset of elements whose lower-case (soft-masked) plus N content is at or below --mask_per percent, and PREFIX.EEs.cleaned.tax.tsv is the taxonomy table restricted to those elements. Both are a subset of the unfiltered pair, which is always written — the flag adds a view rather than replacing one.

The run log (eefinder.log)

eefinder.log is a JSON document recording:

• arguments — the resolved run arguments;

• start_time, end_time, total_time_minutes — wall-clock timing of the run;

• steps_information — one entry per pipeline step, each with its own start/end time, duration and a human-readable message describing what it did with which parameters.

{
    "arguments": {
        "genome_file": "test_files/Ae_aeg_Aag2_ctg_1913.fasta",
        "prefix": "Ae_aeg_Aag2_ctg_1913",
        "outdir": "results_test",
        "mode": "blastx",
        "length": 1000,
        "...": "..."
    },
    "start_time": "2024-10-01 12:00:00",
    "total_time_minutes": "1.8421",
    "steps_information": [
        {
            "step": "Prepare input data",
            "total_time_minutes": "0.0032",
            "message": "Ae_aeg_Aag2_ctg_1913 prefix included in ... sequences header ..."
        }
    ]
}

Intermediate files (tmp_files/)

Unless --removetmp is given, the intermediates are archived under tmp_files/. Their names accrete suffixes as they pass through the pipeline, so you can trace exactly which step produced each file:

tmp_files/
├── PREFIX.rn                        # headers prefixed with PREFIX/
├── PREFIX.rn.fmt                    # contigs below --length removed
├── PREFIX.rn.fmt.fai                # samtools index written by bedtools getfasta
├── PREFIX.rn.fmt.rn.fmt.lenght      # contig lengths, for bedtools slop
├── PREFIX.rn.fmt.blastx             # main similarity search
├── PREFIX.rn.fmt.blastx.csv         # the same hits with the working columns added
├── PREFIX.rn.fmt.blastx.filtred     # redundant-hit filter (--range_junction)
├── PREFIX.rn.fmt.blastx.filtred.bed
├── PREFIX.rn.fmt.blastx.filtred.bed.fasta                    # putative EEs
├── PREFIX.rn.fmt.blastx.filtred.bed.fasta.blastx             # host-bait search
├── PREFIX.rn.fmt.blastx.filtred.bed.fasta.blastx.csv
├── PREFIX.rn.fmt.blastx.filtred.bed.fasta.blastx.filtred
├── ....filtred.concat               # EE + host hits, ranked by bitscore
├── ....concat.nr                    # best hit per element, host winners dropped
├── ....concat.nr.tax                # joined with the -mt metadata
├── ....tax.bed                      # annotated with the --merge_level taxon
├── ....tax.bed.merge                # bedtools merge -d --limit
├── ....merge.fmt                    # annotation stripped back to coordinates
├── ....merge.fmt.fa.bed
└── ....merge.fmt.fa.bed.flank       # coordinates grown by --flank

The three merged-element files that become the main outputs (....merge.fmt.fa, ....merge.fmt.fa.tax and ....merge.fmt.fa.bed.flank.fasta) are renamed into PREFIX.EEs.fa, PREFIX.EEs.tax.tsv and PREFIX.EEs.flanks.fa rather than archived.