Genomes and metagenomes · What you can do here
Check genome assembly completeness online
Assess the completeness of your genome assembly, transcriptome, or gene set with BUSCO 6.1.0 using curated OrthoDB 12 single-copy orthologues, without installing local tools or lineage datasets.
What this involves
BUSCO evaluates sequence completeness by testing for conserved orthologues expected to appear in nearly every species within a lineage as single copies. Rather than relying purely on sequence length, it produces five benchmark figures: Complete (separated into Single-copy and Duplicated), Fragmented, and Missing genes, relative to the total marker count (n). These metrics serve as standard publication quality checks across assemblies and annotations.
Doing this work properly requires matching the execution mode to your data and selecting the most specific lineage available. In genome mode, BUSCO applies an internal gene predictor—Prodigal for prokaryotes or miniprot for eukaryotes—so results reflect both the assembly and the predictor. Proteins mode scores predicted gene sets, while transcriptome mode evaluates expressed sequences. Choosing a specific lineage dataset provides hundreds or thousands of markers, ensuring greater resolution than broad domain-level sets.
The platform's research agent runs BUSCO on its own dedicated machine, managing lineage downloads and returning the summary scores, full coordinate tables, and extracted marker sequences. You supply a FASTA path, URL, or NCBI accession, and the agent monitors the process and reads the results back into your workspace.
How it goes
- 1Supply input sequence and select modeProvide a local FASTA file, download URL, or NCBI accession, then set the mode to genome for assemblies, proteins for annotated gene sets, or transcriptome for transcript collections.
- 2Define the lineage datasetSelect the narrowest applicable OrthoDB 12 lineage dataset, or set the run to auto, auto-prokaryote, or auto-eukaryote to let BUSCO place the sequence taxonomically.
- 3Execute orthologue detectionThe agent initiates BUSCO 6.1.0 as a background job, running Prodigal for prokaryotic genomes or miniprot for eukaryotic assemblies to identify conserved markers.
- 4Extract scores and marker tablesOnce complete, the agent inspects the run directory to extract short summaries, marker coordinates from full_table.tsv, and recovered genes in busco_sequences.
Start it here
Ask in plain language. The agent works on a Linux machine of its own — it installs and runs what the job needs, keeps every file, and reads the results back with the numbers explained. Nothing to set up, and a long run continues after you close the page.
What to watch for
- Comparing BUSCO scores across different OrthoDB generations is invalid because marker sets and thresholds differ between releases.
- Elevated duplication in a haploid or inbred genome usually signals unpurged haplotigs or sample contamination rather than true gene duplication.
- Running proteins mode on a file containing multiple isoforms per gene artificially inflates the Duplicated score unless filtered to the longest protein per gene.
- A high completeness score does not rule out contamination or structural misassemblies, as BUSCO measures only the presence of expected conserved genes.
- The eukaryotic stop-codon metric (E) is rarely zero even in reference assemblies, so compare it against an established relative rather than treating any non-zero value as an error.
What it runs
Programs
- BUSCORuns as a job on its own machine· Software MIT; lineage datasets CC BY-ND 4.0Assesses completeness of genome assemblies, gene sets, and transcriptomes using universal single-copy orthologs.
- SeqKitPreinstalled on the research machineA cross-platform toolkit for fast inspection, filtering, conversion, and analysis of FASTA and FASTQ files.
nf-core pipelines
Questions
Do I need to install anything to check genome assembly completeness?
No. The research agent does this work on a Linux machine of its own — it runs BUSCO, SeqKit there, keeps the files, and reads the results back to you. Long runs continue if you close the page.
Why should I pick a specific lineage instead of a broad domain set?
Broad sets like bacteria_odb12 contain around one hundred markers, meaning each gene alters the score significantly, whereas specific lineages like enterobacterales_odb12 contain hundreds to thousands of markers for much higher diagnostic resolution.
Why do genome and proteins scores differ for the exact same isolate?
Genome mode runs an internal predictor on the raw assembly, while proteins mode scores an existing annotation. A high genome score alongside a lower proteins score indicates that your structural annotation pipeline missed genuine genes present in the assembly.
Does a 98% BUSCO score mean the assembly is finished?
No. BUSCO inspects only conserved single-copy genes, leaving repeats, centromeric regions, and lineage-specific families unmeasured, and it cannot evaluate contiguity (N50) or base-level correctness.
How should I handle assemblies of polyploid organisms?
True whole-genome duplications naturally increase the Duplicated metric. In clades with polyploidy, a high duplication figure reflects genuine biology rather than unpurged heterozygous contigs, provided the organism's history supports it.
Written from the method the research agent follows for this work and the documentation of the tools listed above. Statements citing a figure that does not appear there are removed before publication. Corollary is a non-commercial research demonstration; results are not for clinical use.