SeqChem

Quick start

Convert a SMILES string to InChI:

curl -s https://seqandchem.com/api/v1/convert \
  -H 'Content-Type: application/json' \
  -d '{"from":"smi","to":"inchi","data":"CC(=O)Oc1ccccc1C(=O)O"}'

Same call from Python:

import requests

r = requests.post("https://seqandchem.com/api/v1/convert", json={
    "from": "smi",
    "to": "inchi",
    "data": "CC(=O)Oc1ccccc1C(=O)O",
})
r.raise_for_status()
print(r.json()["result"])
# InChI=1S/C9H8O4/c1-6(10)13-8-5-3-2-4-7(8)9(11)12/h2-5H,1H3,(H,11,12)

Or use the official client — standard library only, no dependencies:

pip install seqchem

from seqchem import Client

c = Client()
print(c.convert("CC(=O)Oc1ccccc1C(=O)O", to="inchi")["result"])
print(c.properties("CCO")[0]["props"]["MW"])      # 46.069
print(c.similarity("CCO", "CCCO"))                # Tanimoto

And from the browser — CORS is enabled, so this works from any origin:

const res = await fetch("https://seqandchem.com/api/v1/properties", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({ from: "smi", data: "CC(=O)Oc1ccccc1C(=O)O" }),
});
const data = await res.json();
console.log(data.rows[0].props.MW);   // 180.159

Format identifiers

Every format is referred to by its OpenBabel id: smi, can, inchi, inchikey, mol, sdf, mol2, pdb, xyz, cml, smi, png, svg, pdf and roughly 180 more. Get the full list with GET /api/v1/formats — it also tells you which formats can be read and which can be written.

curl -s https://seqandchem.com/api/v1/formats | python3 -m json.tool | head -20

Endpoints

All paths below are relative to https://seqandchem.com. Every POST endpoint accepts a JSON body and returns JSON.

Method	Path	Purpose
GET	/api/v1/formats	List supported formats
GET	/api/v1/example	Example molecule in any format
POST	/api/v1/convert	Convert between formats
POST	/api/v1/properties	Molecular descriptors
POST	/api/v1/depict	2D structure as SVG
POST	/api/v1/gen3d	3D conformer as SDF
POST	/api/v1/substructure	SMARTS substructure filter
POST	/api/v1/fingerprint	FP2/FP3/FP4/MACCS fingerprint
POST	/api/v1/similarity	Tanimoto similarity
POST	/api/v1/iupac	IUPAC name → structure
POST	/api/v1/batch	Bulk file conversion
POST	/api/v1/formula	Formula ↔ structure
POST	/api/v1/bio	FASTA / FASTQ operations

GET /api/v1/formats

Lists every format with a description and whether it can be read, written, or both.

Query	Type	Description
dir	string	Optional filter: read or write

{
  "ok": true,
  "items": [
    {"id": "smi", "desc": "SMILES format", "read": true, "write": true},
    {"id": "inchi", "desc": "InChI format", "read": true, "write": true}
  ]
}

GET /api/v1/example

Returns aspirin in the requested format. Useful to check what a format looks like before writing a parser.

Query	Type	Description
format	string	Format id, default smi

{"ok": true, "format": "mol", "data": " aspirin\n ..."}

POST /api/v1/convert

Converts one or more records between any two readable/writable formats. Text output comes back in result; binary output (PNG, PDF, STL) comes back base64-encoded in data_url.

Field	Type	Description
data	string	Required. Molecule data in the source format
from	string	Input format id, default smi
to	string	Output format id, default mol
opts	string[]	Optional OpenBabel flags: --gen3d, --canonical, --addhydrogens, --deletehydrogens, --unique, --separate

{
  "ok": true,
  "binary": false,
  "result": "InChI=1S/C9H8O4/c1-6(10)13-8-5-3-2-4-7(8)9(11)12/h2-5H,1H3,(H,11,12)",
  "mime": "text/plain",
  "is_image": false,
  "log": "1 molecule converted"
}

Requesting a 3D format (sdf, mol2, xyz, pdb) adds --gen3d automatically if you do not pass it.

POST /api/v1/properties

Computes descriptors for every record in data. One row per molecule.

Field	Type	Description
data	string	Required. One or more molecules
from	string	Input format id, default smi

{
  "ok": true,
  "rows": [{
    "smiles": "CC(=O)Oc1ccccc1C(=O)O",
    "props": {
      "formula": "C9H8O4", "MW": "180.159", "logP": "1.3101",
      "TPSA": "63.6", "HBD": "1", "HBA1": "4", "rotors": "3",
      "atoms": "21", "bonds": "21", "MR": "44.691",
      "cansmi": "CC(=O)Oc1ccccc1C(=O)O",
      "InChI": "InChI=1S/C9H8O4/c1-6(10)...",
      "InChIKey": "BSYNRYMUTXBXSQ-UHFFFAOYSA-N"
    }
  }],
  "labels": {"formula": "Formula", "MW": "Molecular weight"}
}

POST /api/v1/depict

Renders a 2D depiction and returns the raw SVG markup in svg. The SVG is self-contained — embed it directly in a page.

Field	Type	Description
data	string	Required. Molecules to draw
from	string	Input format id, default smi

{"ok": true, "svg": "<svg xmlns=\"http://www.w3.org/2000/svg\" ...></svg>"}

POST /api/v1/gen3d

Generates a force-field optimised 3D conformer and returns an SDF block. This is a heavy operation: it is serialised across the whole server, so under load you may get 429 with a retry hint.

Field	Type	Description
data	string	Required. Molecules
from	string	Input format id, default smi

{"ok": true, "result": " aspirin\n     RDKit          3D\n..."}

POST /api/v1/substructure

Filters molecules by a SMARTS pattern and returns the matching ones.

Field	Type	Description
data	string	Required. Molecules to search
smarts	string	Required. SMARTS pattern, max 500 chars
from	string	Input format id, default smi

{"ok": true, "count": 2, "matches": ["CC(=O)Oc1ccccc1C(=O)O", "CC(=O)Nc1ccc(O)cc1"]}

POST /api/v1/fingerprint

Returns the raw fingerprint bit string for each molecule.

Field	Type	Description
data	string	Required. Molecules
fp	string	FP2 (default), FP3, FP4, MACCS, ECFP2, ECFP4, ECFP6
from	string	Input format id, default smi

{"ok": true, "result": "00000000 00000000 00000001 ...", "words": 18}

POST /api/v1/similarity

Tanimoto coefficient between two molecules.

Field	Type	Description
a	string	Required. First molecule
b	string	Required. Second molecule
fp	string	FP2 (default), FP3, FP4, MACCS
from	string	Input format id, default smi

{"ok": true, "tanimoto": 0.4815, "bits_a": 54, "bits_b": 52, "common": 25, "union": 81}

POST /api/v1/iupac

Parses an IUPAC name with OPSIN (Cambridge) and returns the structure identifiers.

Field	Type	Description
name	string	Required. IUPAC name, max 300 chars

{
  "ok": true,
  "resolved": true,
  "source": "opsin",
  "smiles": "CC(=O)Oc1ccccc1C(=O)O",
  "inchi": "InChI=1S/C9H8O4/c1-6(10)...",
  "inchikey": "BSYNRYMUTXBXSQ-UHFFFAOYSA-N"
}

If the name cannot be interpreted you get ok: true, resolved: false with a message — this is not an error, just a miss.

POST /api/v1/batch

Bulk conversion of a whole file. Returns result for text formats; for image formats it returns a per-molecule preview list and, when there is more than one record, a ZIP as a zip_url data URL. Limits: 64 MB input, 5,000 records per run (500 for image output).

Field	Type	Description
data	string	Required. Multi-record file
from	string	Input format id, default sdf
to	string	Output format id, default smi
opts	string[]	--canonical, --unique, --deletehydrogens, --addhydrogens

{"ok": true, "binary": false, "result": "CC(=O)Oc1ccccc1C(=O)O\n...", "converted": 250}

POST /api/v1/formula

Two modes: parse a formula string, or compute formula and weight from a structure.

Field	Type	Description
mode	string	formula (default) or structure
formula	string	Used when mode is formula, e.g. C9H8O4
data	string	Used when mode is structure
from	string	Input format id for structure mode

{"ok": true, "formula": "C9H8O4", "mw": 180.159,
 "composition": {"C": 9, "H": 8, "O": 4}, "atoms": 21}

POST /api/v1/bio

Sequence operations, all implemented natively in Python.

Field	Type	Description
op	string	Required. fasta2fastq, fastq2fasta, revcomp, translate, stats
data	string	Required. FASTA, FASTQ or a bare sequence
frame	integer	Reading frame for translate, 0–2

{"ok": true, "result": "ATGGCC...", "count": 3}

Errors

Failures always return the same shape, with an HTTP status of 400, 422 or 429:

{"ok": false, "error": "Conversion failed", "detail": "1 molecule converted"}

Status	Meaning
400	Bad request — missing field, unknown format, input too large
422	OpenBabel or OPSIN could not process the input
429	Rate limit or a heavy job already running; see Retry-After
502	Upstream name-resolution service unavailable

Limits and fair use

• 120 requests per minute per client, 300 per minute across the whole service.
• 8 MB maximum input per request; 64 MB for /api/v1/batch.
• 5,000 records per batch run, or 500 when the output is an image (PNG, SVG, PDF…). Requests over the limit return 400 with the actual count and the maximum.
• 3D conformer generation is serialised — one at a time across all users. Plan for a few seconds per molecule.
• If you need sustained bulk throughput, generate 3D conformers locally with OpenBabel or RDKit; this API is meant for interactive and small-batch use.

Attribution

The service is free to use. Chemistry is powered by OpenBabel 3.1.1 and name parsing by OPSIN; if you publish something built on this API, a link back to seqandchem.com is appreciated but not required.

Machine-readable specification: /api/openapi.json — import it into Postman, Insomnia, or any OpenAPI client generator.

Questions, bug reports or a use case the API does not cover: devin2xing@gmail.com.