BGPT: Paper Review: TeCD: The eccDNA Collection Database for extrachromosomal circular DNA

Fuel Your Discoveries

Quick Explanation Copied

TeCD (eccDNA Collection Database) — what it adds

TeCD compiles cross-species eccDNA (animals, plants, fungi) into standardized records with genomic coordinates, gene/transposon context, and an in-database BLAST interface for retrieval and similarity search.

Long Explanation

Paper review (skeptical, evidence-based): TeCD — eccDNA Collection Database

Citation:

1) Visual first: what TeCD actually claims (quantified)

TeCD reports a multi-step pipeline: from an estimated ~2,343,000 literature-reported eccDNA candidates to 1,846,905 after coordinate-based filtering, then 948,469 after BLAST-based deduplication (e-value threshold ≤ 1e-6).

2) Cross-species composition: before/after BLAST deduplication

TeCD gives species-wise counts both before and after BLAST deduplication.

3) Methods summary (what pipeline steps mean biologically)

Literature extraction → coordinate-based filtering. The paper argues eccDNA “comes from the chromosome” and therefore keeps only records with extractable genomic start/end sites mapped to a corresponding reference genome.
Cross-species standardization. TeCD processes records for five eukaryotes (Homo sapiens, S. cerevisiae, A. thaliana, G. gallus, M. musculus) and assigns consistent identifiers using a species code + ordering.
Deduplication by sequence similarity. The paper removes repeated sequences within species using BLAST with an e-value threshold of 1e-6.
Gene and transposable element context. TeCD links eccDNA coordinates to gene annotations (NCBI Gene; TAIR for Arabidopsis) and links transposable elements to Dfam for repetitive-element interpretation.
In-database BLAST interfaces. The paper reports BLASTN, TBLASTN, and TBLASTX support inside TeCD, which enables sequence-to-record similarity search against TeCD’s eccDNA set.

4) Core value to a scientist (and what it’s not)

What TeCD is well-positioned to do:

Retrieval of eccDNA genomic context (host locus coordinates, overlap with genes, overlap with transposons, and the sequencing approach/source PMIDs).
Similarity-based cross-study reuse via TeCD’s BLAST implementation, enabling users to ask “what previously reported eccDNAs look like this sequence (or translate similarly)?”
Comparative statistics across species, sample/tissue/cell categories, and chromosome-level distribution (the paper describes histograms and Z-score standardization of eccDNA-per-chromosome-length).

What TeCD is not (from what’s provided here):

Not a mechanistic validation engine. The entries originate from published literature; TeCD standardizes and deduplicates, but the paper does not claim experimental re-validation of every record.
Not a uniform experimental harmonization across technologies. Because TeCD includes multiple eccDNA isolation/sequencing approaches across studies, any cross-record comparisons inherit heterogeneity in upstream detection conditions.

5) Skeptical critique: key limitations & possible failure modes

Limitation A — coordinate reliability depends on extraction from papers. TeCD keeps only records with start/end sites and maps them to reference genomes; if start/end coordinates are imprecise or inconsistently reported across sources, mapping accuracy propagates into TeCD. Limitation B — deduplication may remove biologically distinct circles. Deduplication is sequence-similarity based within species (e-value ≤ 1e-6). If distinct eccDNAs share highly similar or identical sequences (e.g., from different contexts, cells, tissues, or structural forms), they may collapse into fewer records. Limitation C — cross-study detection bias. TeCD includes eccDNAs enriched under particular conditions (e.g., the paper mentions a zeocin vs control analysis in yeast), but global comparisons across tissues/samples may still reflect differences in library preparation and selection. Limitation D — reference genome and annotation versioning. TeCD maps to specific reference genomes (e.g., GRCh38.p13, S288C, TAIR10.1, GRCg7b, GRCm39). As reference/annotation updates occur, coordinate interpretation can shift. Blind spot — “biological function” claims are constrained by database nature. The paper describes analyzing possible potential functions of eccDNA, but TeCD’s core deliverable is collection + annotation + search; functional inferences will be limited by which genomic features are captured and by annotation completeness.

6) One concrete in-paper analysis example (zeocin vs control)

TeCD reports a yeast example using data with and without the DNA damaging agent zeocin. It states the control group had an average of 147 eccDNAs per group, while the zeocin group had an average of 219; it further reports bidirectional BLAST comparisons yielding that matched eccDNA sequences accounted for ~11–13% of average eccDNAs in both conditions.

7) Why this matters (confidence-weighted)

High confidence: TeCD’s strongest contribution is infrastructural—standardized eccDNA records across five eukaryotes with BLAST-enabled retrieval, plus cross-links to gene/TE resources and downloadable data.
Moderate confidence: The biological inferences enabled by TeCD (e.g., gene overlap, TE enrichment, condition associations) are plausibly useful but remain bounded by literature heterogeneity and coordinate/dedup assumptions.

Author-specific deep dive links (BGPT)

Feedback:

Updated: April 02, 2026