https://github.com/vmikk/eutax
Backend for taxonomic annotation UI
https://github.com/vmikk/eutax
Last synced: 6 months ago
JSON representation
Backend for taxonomic annotation UI
- Host: GitHub
- URL: https://github.com/vmikk/eutax
- Owner: vmikk
- License: apache-2.0
- Created: 2025-04-14T13:33:51.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2025-05-23T17:42:40.000Z (8 months ago)
- Last Synced: 2025-05-23T18:43:51.999Z (8 months ago)
- Language: Python
- Homepage:
- Size: 239 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Taxonomic Annotation API
A FastAPI-based server for taxonomic annotation of DNA sequences.
## Features
- Upload FASTA files containing DNA sequences
- Run taxonomic annotation using BLAST or VSEARCH
- Track job status and retrieve results
- RESTful API with auto-generated documentation
## Results Format
The `results.json` file contains taxonomic annotation results in a structured format.
The exact content depends on the tool used (BLAST or VSEARCH), but follows this general structure:
```json
{
"results": [
{ ... }, # matches for each query sequence
{ ... },
...
],
"summary": {
"total_queries": 4,
"total_hits": 80
},
"metadata": {
"tool": "blast",
"algorithm": "megablast",
"database": {
"identifier": "eukaryome_its",
"version": "1.9.4"
},
"job_id": "job123"
}
}
```
For each query sequence, the results are stored in the `results` array. For example:
``` json
{
"results": [
{
"query_id": "query_name",
"query_length": 553,
"hit_count": 2,
"hits": [
{
"sseqid": "EUK1101818;Fungi;Basidiomycota;Tremellomycetes;Filobasidiales;Piskurozymaceae;Solicoccozyma;aeria",
"taxonomy": {
"accession": "EUK1101818",
"kingdom": "Fungi",
"phylum": "Basidiomycota",
"class": "Tremellomycetes",
"order": "Filobasidiales",
"family": "Piskurozymaceae",
"genus": "Solicoccozyma",
"species": "aeria"
},
"pident": 100.0,
"length": 553,
"mismatch": 0,
"gapopen": 0,
"qstart": 1,
"qend": 553,
"sstart": 1235,
"send": 1787,
"evalue": 0.0,
"bitscore": 998.0,
"qcovs": 100.0,
"sstrand": "plus",
"slen": 4063,
"alignment": {
"qseq": "GTGGGATTAAA...", # truncated
"midline": "||| |||||...", # truncated
"sseq": "GTGAATTAAA..." # truncated
}
},
{ ... }, # other hits
]
},
{ ... }, # other query sequences
{ ... }
],
"summary": { ... },
"metadata": { ... }
}
```
Each hit contains the following information:
- **sseqid**: Identifier of the reference sequence
- **taxonomy**: Taxonomic classification object with fields (accession, kingdom, phylum, class, order, family, genus, species)
- **pident**: Sequence identity percentage
- **length**: Length of the alignment in nucleotides
- **mismatch**: Number of mismatched positions
- **gapopen**: Number of gap openings in the alignment
- **qcovs**: Percentage of query covered by alignment
- **evalue**: E-value (not available for VSEARCH output)
- **bitscore**: Bit score (for VSEARCH, raw score is used)
- **qstart** and **qend**: Start and end positions in query
- **sstart** and **send**: Start and end positions in subject/target
- **sstrand**: Alignment strand (e.g., "plus" or "minus")
- **slen**: Length of the subject sequence
- **alignment**: Object containing aligned sequences with fields:
- **qseq**: Aligned portion of query sequence
- **midline**: Match representation between sequences
- **sseq**: Aligned portion of subject sequence