SMILES to InChI
v1.5Generate InChI identifiers from SMILES strings for registration and interoperability workflows.
Generate InChI identifiers from SMILES strings for registration and interoperability workflows.
SMILES to InChI is a small-molecule identifier converter that turns a SMILES molecular graph into an IUPAC International Chemical Identifier, or InChI. SMILES is concise and convenient for storing or drawing a structure, but equivalent molecules can be written as different valid SMILES strings. An InChI applies a standardized structure-normalization procedure and writes the result as a layered identifier, making it useful for exchanging well-defined chemical structures across databases and software.
The conversion is performed with RDKit and its InChI support. Each submitted SMILES is parsed and chemically sanitized before the InChI is written. Invalid syntax, impossible valences, or structures RDKit cannot sanitize are reported as failed entries rather than being exported as uncertain identifiers.
An InChI carries structural information such as formula, connectivity, hydrogens, charge, and, when specified and retained, stereochemistry and isotopic composition. It is a structure identifier, not a 3D conformer or a prediction of the dominant protonation state at a particular pH. For a structure file with coordinates, SMILES to SDF is the appropriate next step.
ProteinIQ converts one to ten SMILES entries into downloadable InChI identifiers. Paste one SMILES per line, upload a supported text file, or retrieve compounds from PubChem by name or CID. Each successfully parsed molecule receives an individual .inchi file, while multi-compound jobs also include a tab-separated batch file and a curated Run log that records failed inputs.
| Input | Accepted form | Notes |
|---|---|---|
SMILES text | One SMILES per line | For example, ethanol is CCO and aspirin is CC(=O)Oc1ccccc1C(=O)O. Empty lines and comment lines beginning with # are ignored. |
| Named batch | name<TAB>SMILES | Use a literal tab between the label and SMILES. The label is retained in the batch output and used to form the individual filename. |
| File upload | .txt, .csv, .tsv, .smi, or .smiles | The submitted text is read as molecular entries, up to 10 entries per job. |
| PubChem retrieval | Compound name or CID | ProteinIQ retrieves the corresponding SMILES before applying the same conversion. Confirm the resolved structure when a name can refer to more than one chemical form. |
A named input can be written as:
aspirin CC(=O)Oc1ccccc1C(=O)OSMILES should describe the intended chemical form before submission. In particular, charge, explicit stereochemistry, salts, isotopes, and tautomer choice can all affect the resulting InChI. A compound name is not accepted as plain SMILES text; use PubChem retrieval when a name or CID is the available starting point.
| Output | When it appears | Contents |
|---|---|---|
Per-compound .inchi file | Each successful entry | One full InChI string followed by a newline. The filename is based on the submitted label. |
batch.inchi.tsv | Two or more entries convert successfully | Tab-separated name and InChI values for all successful entries. |
run.log | Every job | A curated run record with the input count, conversion phase, successful and failed entry counts, and stable failure guidance. |
One failed structure does not stop the remainder of the batch. Failed entries are omitted from the .inchi files and batch.inchi.tsv, while the Run log records the failed-entry count.
The converter first reads the SMILES into an RDKit molecular graph. RDKit sanitization checks the graph for chemically consistent valence and assigns features such as aromaticity. The tool then removes hydrogens that can be represented implicitly and asks the InChI library to write the resulting structure as an InChI.
The identifier commonly begins with InChI=1S/, where 1S denotes a Standard InChI produced with the standard settings. Sections separated by / encode different layers of structural information. For example, the ethanol SMILES CCO produces InChI=1S/C2H6O/c1-2-3/h3H,2H2,1H3.
The tool does not generate an InChIKey. An InChIKey is a fixed-length hash derived from an InChI and is useful for indexing and web search, but it cannot be decoded back into a molecular structure. A full InChI can be converted back to a toolkit-specific canonical SMILES with InChI to SMILES.
The InChI is a normalized identifier for the molecular structure RDKit and the InChI library obtained from the submitted SMILES. It should be treated as a chemistry-aware matching aid, not as a byte-for-byte preservation of the input notation.
SMILES to InChI fits registry, deduplication, and data-exchange workflows where a standardized structure identifier is needed. InChI to SMILES is the reverse conversion for workflows that require a SMILES string. Neither direction should be used as a literal round-trip test because SMILES and InChI apply different representation and normalization rules.
For conversions involving coordinates, atom properties, or formats beyond a compact identifier, use Open Babel. It can translate between a wider range of chemical and structure file formats, whereas this tool focuses on producing a full InChI from SMILES.
Generate SMILES strings from InChI identifiers for search, filtering, and modeling workflows.
Extract canonical SMILES strings from MDL MOL ligand files for cheminformatics workflows.
Extract canonical SMILES strings from MOL2 ligand files for cheminformatics workflows.
Convert ligand and peptide PDB coordinates to SDF with RDKit bond perception.
Export ligand poses from Meeko-prepared PDBQT to SDF while recovering chemical connectivity.
Convert SDF ligands to PDBQT with Meeko, preserving supplied coordinates.
Extract canonical SMILES strings from SDF ligand files, one record at a time or in batch.
Generate MOL2 files from SMILES strings for docking, scoring, and modeling workflows.
Generate 3D PDB files from SMILES strings for visualization and small-molecule workflows.
Generate 3D ligand coordinates from SMILES and prepare PDBQT files with Meeko.