AtomLens

What it does

• Open chemistry formats such as MOL, SDF, SMILES, CXSMILES, Reaction SMILES, InChI, MOL2, PDB, XYZ, CML, RXN, and RDF.
• Render clean 2D molecule views and dedicated reaction views in a native macOS interface.
• Generate identifiers: SMILES, isomeric SMILES, InChI, and InChI Key.
• Calculate core properties like formula, molecular weight, monoisotopic mass, LogP, and Rule-of-Five status.
• Split multi-structure SDF and multi-reaction SMILES files into individual outputs for downstream work.
• Inspect advanced chemistry notation such as supported Markush R-groups, imported highlights, and reaction-aware CML input.
• Search molecule and reaction files locally through Spotlight metadata and preview them with Quick Look.

Tip: The Help page includes practical search queries and full format tables.

Designed for local workflows

AtomLens 1.3 deepens day-to-day chemistry workflows with broader format fidelity, CXSMILES and Markush support, reaction-aware CML import, and smoother ways to inspect, extract, and organize local files. The focus stays the same: clear structure views, practical outputs, and everything staying on your Mac.

What's new?

Version 1.3 makes AtomLens more capable with modern chemistry notation, richer roundtrip fidelity, and more practical dataset handling on macOS, without making the app harder to use.

• Open dedicated .cxsmiles files and depict supported CXSMILES features including Markush R-groups, repeat and link-node notation, and imported highlights.
• Improved Markush rendering keeps the main scaffold readable while R-group legends stay clearly separated below.
• Better import and export fidelity for MDL V2000, V3000, SDF, and CML workflows, preserving more of the chemistry that matters in day-to-day use.
• Reaction handling now extends to reaction-bearing CML files alongside Reaction SMILES, RXN, and RDF.
• New CSV extraction turns annotated multi-structure SDF datasets into analysis-ready tables for downstream work.
• Expanded highlight display options make annotated and query-rich structures easier to inspect at a glance.

Try sample structures

Use this starter set of molecule and reaction files to test parsing, rendering, extraction, and Spotlight/Quick Look workflows in AtomLens.

Download example structures (ZIP)

Stable link for future site updates: /assets/downloads/atomlens-example-structures.zip

Screenshots

The spark: why this app exists

If you work with molecules, you know the moment: someone sends "the compound" as an SDF attachment, your downloads folder becomes a museum, and a week later you remember it had an InChI Key — but not the filename. Traditional cheminformatics tools are powerful, but the operating system itself still treats molecule files like anonymous text.

AtomLens started as a simple wish: "Finder should be able to preview and search molecules like it previews and searches PDFs." From that, a surprisingly ambitious requirements list fell out:

• Open common chemistry formats: SDF, MOL, SMILES, InChI, and more.
• Render chemist-friendly 2D depictions: aromatic rings read as aromatic, chains are readable, stereo bonds are unambiguous.
• Compute identifiers and properties: SMILES, ISO SMILES, InChI, InChI Key, formula, molecular weight, LogP, Rule of Five.
• Index those values into Spotlight and show them in Finder via Quick Look previews and thumbnails.
• Remain App Store compliant: sandbox-safe, predictable permissions, and no external converter dependencies.

Spotlight, Quick Look, and the macOS "superpowers"

This is the part that makes AtomLens feel "native" in a way most scientific tools do not: structures become searchable and previewable without opening the app. Spotlight indexing adds metadata to molecule and reaction files (SMILES, InChI, InChI Key, formula, weight, LogP, and reaction flags), and Quick Look turns Spacebar into an instant chemistry preview.

We also built a Molecule Browser that queries Spotlight directly. It is essentially a local chemistry-library UI without a separate database - just the macOS index you already have.