GenBank to GFF3/GFF2 Converter

Used this tool in your research?

Cite it instantly and export to BibTeX, EndNote, RIS, and more.

Generate Citation

The ScienceCodons GenBank to GFF3/GFF2 Converter is a free online bioinformatics tool for converting annotated GenBank files into GFF3 (General Feature Format version 3) or GFF2 (General Feature Format version 2). It transforms GenBank sequence annotations into structured, nine-column files while preserving genomic coordinates, strand information, gene identifiers, and important feature relationships. Designed for researchers working with genome annotations, gene structures, and comparative genomics, the converter makes it easier to prepare annotation files for genome browsers and downstream bioinformatics workflows without manually rewriting feature tables or running conversion scripts. If your analysis specifically requires Gene Transfer Format for transcriptomics, you can also use our GenBank to GTF Converter.

How to Convert GenBank to GFF3 or GFF2

Converting a GenBank record takes just a few steps:

1. Upload your GenBank file: Select the “Upload File” tab and drag and drop your file into the upload area, or click to browse your device. Supported GenBank file extensions include .gb, .gbk, and compatible text files.
2. Or paste your GenBank data: Choose the “Paste Text” tab to insert the complete contents of an annotated GenBank record directly into the text field.
3. Select your output format: Choose GFF3 or GFF2 from the “Target Format” menu. GFF3 is selected by default on this page.
4. Run the conversion: Click “Convert Format” to process the annotations and generate the selected output format.
5. Preview and download: Review the converted annotations in the results panel. You can copy the output or download it as a .gff3 or .gff2 file.

Why Convert GenBank to GFF3?

GenBank files combine nucleotide sequences with detailed biological annotations, including gene locations, coding regions, RNA features, protein products, and database identifiers. This information is valuable for research, but its flat-file format isn’t always convenient for applications that expect structured genomic features.

GFF3 provides a standardized representation in which genes, transcripts, exons, and coding sequences can be described as individual features with explicit relationships.

Converting GenBank annotations to GFF3 can help researchers with:

• Genome visualization: Prepare gene and feature annotations for compatible genome browsers, including JBrowse and IGV.
• Gene structure analysis: Examine the organization of genes, transcripts, exons, and coding sequences.
• Comparative genomics: Standardize annotation records when comparing genomic regions across organisms or reference assemblies.
• Annotation integration: Prepare existing GenBank records for pipelines and software that accept GFF3 input.
• Genomic feature inspection: Work with genomic coordinates, feature identifiers, and annotation attributes in a structured format.

The converter is particularly useful when annotated sequences are already available in GenBank format, but the next stage of analysis requires GFF3 or GFF2.

Key Features

1. Gene and Transcript Hierarchy Reconstruction

One of the most useful features of GFF3 is its ability to represent relationships between biological features.

For example, a gene may contain an mRNA transcript, which in turn contains exons and coding sequence (CDS) features.

ScienceCodons organizes these relationships using GFF3 attributes such as ID and Parent, making it easier for compatible software to interpret the structure of annotated genes.

When relationships are not explicitly defined in the original GenBank record, the converter uses available identifiers and feature locations to associate related annotations. The resulting hierarchy depends on the information available in the input record.

2. Support for Complex Genomic Locations

GenBank records can describe genes located on the forward or reverse DNA strand, as well as features divided into multiple genomic regions.

The converter interprets common GenBank location expressions, including join(), complement(), and order(), to represent supported features in the selected output format.

This is particularly important for genes containing multiple exons or coding segments.

3. Accurate Coordinates and Strand Orientation

The converter preserves the 1-based, inclusive genomic coordinate system used by GenBank and GFF formats.

For features located on the reverse strand, the output records the strand as – while maintaining genomic start and end coordinates in the expected order.

4. Coding Sequence Phase Calculation

For protein-coding genes, accurate CDS phase information is important for interpreting coding regions that span multiple segments.

The converter calculates CDS phase values using the available coding sequence information, including GenBank’s codon_start qualifier and the arrangement of coding segments.

These calculations help maintain a consistent representation of protein-coding annotations in the converted file.

5. Preservation of Relevant Annotation Details

GenBank annotations can contain useful biological descriptors such as gene names, locus tags, protein identifiers, and product descriptions. The converter carries relevant information into the corresponding GFF attributes where supported, helping researchers retain important feature context. GFF3-specific characters are also encoded appropriately to avoid disrupting attribute parsing.

6. GFF3 and Legacy GFF2 Output

In addition to GFF3, the converter supports GFF2 for workflows that require the older annotation format.

You can select the appropriate format without switching to another converter or manually modifying the output.

GFF3 vs. GFF2

Although GFF3 and GFF2 share a nine-column structure, they use different conventions for representing annotation attributes and feature relationships.

FeatureGFF3GFF2

File extension .gff3 .gff2

Structure Nine tab-separated columns Nine tab-separated columns

Attributes Structured tag=value pairs Older attribute syntax

Feature relationships Explicit ID and Parent relationships Less explicitly standardized hierarchy

Typical use Modern genome annotations and compatible genome browsers Legacy annotation tools and pipelines

Choose GFF3 if you need structured relationships between genes, transcripts, exons, and coding sequences or are preparing annotations for modern genome analysis applications.

Choose GFF2 if your software specifically requires the older General Feature Format specification.

For gene expression analysis and RNA-seq workflows that expect gene_id and transcript_id attributes, consider the GenBank to GTF Converter instead.

Always check the required annotation format and conventions of your downstream software before importing converted files.

Understanding the GFF3 Output Structure

GFF3 represents each genomic feature as a separate row containing nine tab-separated columns.

These columns specify the sequence identifier, annotation source, feature type, start position, end position, score, strand, CDS phase, and feature attributes.

The ninth column is especially important because it describes the identity of a feature and its relationship to other features.

For example, a gene and its transcript may be represented using attributes such as:

Gene:

ID=gene-YJR064W;Name=TCP1

mRNA:

ID=rna-YJR064W;Parent=gene-YJR064W

Exon:

ID=exon-YJR064W-1;Parent=rna-YJR064W

These attributes show how individual features belong to the same annotated gene structure.

The Parent attribute connects a child feature to its corresponding parent, allowing downstream applications to reconstruct biological relationships.

Depending on the input record, the generated GFF3 file may include gene, mRNA, exon, CDS, region, and other supported feature types.

The conversion reorganizes existing annotations and may infer relationships from available feature information. It does not independently predict new genes or experimentally validate transcript structures.

Scientific Validation and Benchmark Results

The ScienceCodons GenBank annotation converter was evaluated through 16 targeted test scenarios covering GFF3, GFF2, and GTF conversion behavior.

Benchmark summary

• 16 out of 16 tests passed across the shared conversion engine, including expected rejection of unsupported inputs.
• GFF3 hierarchy validation: An annotated Saccharomyces cerevisiae GenBank record (NCBI accession U49845) successfully passed the gene hierarchy test, with a recorded response time of 713.9 milliseconds.
• GFF2 compatibility: The dedicated legacy GFF2 formatting test passed, with a recorded response time of 479.1 milliseconds.
• Complex annotation handling: Tests covered multi-exon structures, reverse-strand genes, CDS phase calculations, non-coding RNA features, pseudogenes, and partial genomic coordinates.
• Response-time testing: A separate series of five consecutive requests recorded an average response time of approximately 476 milliseconds.

The benchmark also examined how the converter responds to problematic inputs, including malformed or unsupported genomic structures.

These tests provide evidence of successful behavior under the evaluated conditions. They do not establish universal compatibility with every GenBank annotation or guarantee the same response time for all files.

Processing time may vary with sequence length, annotation complexity, and network conditions.

Input Requirements and Conversion Limitations

The converter is intended for annotated GenBank flat files rather than raw sequence data. Each input should contain a properly formatted GenBank record with a LOCUS header, a FEATURES section, and the closing // record marker.

A few limitations are important to understand before conversion:

• Single-record processing: The online converter currently supports one GenBank record per conversion. Files containing multiple separate records must be split and processed individually.
• Circular genome boundaries: Some features in circular genomes extend across the coordinate origin. These origin-crossing features are not currently supported. The converter detects such cases and reports an error rather than generating potentially misleading coordinates.
• Annotation completeness: Missing or ambiguous gene and transcript information in the original record may affect how feature relationships are represented.
• Large files: Upload and processing limits may affect very large genomic records. For substantial datasets or high-volume conversion requirements, please contact ScienceCodons to discuss available options.

For research involving critical gene models, we recommend reviewing the converted annotation before incorporating it into downstream analyses.

Data Privacy and Security

We recognize that genomic datasets may include unpublished research, experimental results, or other confidential scientific information. ScienceCodons processes GenBank files through an encrypted connection. Submitted data is handled in memory during conversion, and the conversion service does not save uploaded sequences or generated GFF files to persistent storage.The converted output is returned directly to your browser, where you can preview, copy, or download it.