PubChem advanced search: molecular weight, heavy atoms, HBA

This page shows how to run PubChem searches by name, identifier, structure, or property, from single-compound lookup to batch queries. Matching the search mode to the data you have — a specific CID or InChIKey for fast, exact lookups, or a name plus property filters for broader exploration — is what most affects how relevant and manageable your results are.

What you can ask for

You run these searches by asking Ivy in chat; each runs inside that turn, with no job to track.

Example prompts

Look up aspirin, ibuprofen and caffeine in PubChem
Get the PubChem properties for CID 2244
Find PubChem compounds at least 80% similar to CC(=O)NC1=CC=CC=C1C(=O)O
Find PubChem compounds containing a benzimidazole substructure

The web app also has a PubChem Database Search form in the Tools panel. It submits a PubChem job (compound identification, comprehensive search, fragment library generation, similarity analysis, or batch processing of CIDs) that you can track in your job history; the same job can be submitted through the REST API or ivybloom run. The chat searches above are a separate, faster path.

PubChem search strategies in .bloom* platform

Choose search strategies based on your available data and research objectives. .bloom*'s PubChem integration provides multiple search modes optimized for different use cases.

Interactive search methods

Text and name-based search

Chemical name search:

• Common names: "aspirin", "caffeine", "ibuprofen"
• Systematic names: IUPAC nomenclature
• Brand names: Pharmaceutical trade names
• Synonyms: Alternative chemical designations

Search tips:

• Use common names for drug-like compounds
• Try multiple synonyms if initial search yields no results
• Combine with property filters to narrow ambiguous results

Identifier-based search

PubChem-specific identifiers:

• CID: Compound ID (e.g., 2244)
• SID: Substance ID for different forms/suppliers
• AID: Assay ID for bioactivity data linkage

Standard chemical identifiers:

• CAS numbers: Chemical Abstracts Service registry numbers
• InChIKey: Hashed InChI for exact structure matching
• SMILES: Chemical structure notation
• InChI: IUPAC International Chemical Identifier

Best practices:

• Use CID for fastest, most reliable lookups
• Prefer InChIKey for exact structure verification
• CAS numbers work well for regulatory compounds

Structure-based search

SMILES input:

• Direct SMILES string entry with validation
• Support for stereochemistry and complex structures
• Automatic format detection and standardization

Substructure searching:

• SMARTS pattern matching for functional groups, via a PubChem substructure search
• Scaffold-based queries for analog series

Advanced search strategies

Property-constrained searching

PubChem's search does not take property ranges as input. Ask Ivy for the filters below and it applies them to the properties PubChem returns, or with RDKit in the chat's code workspace for a larger set.

Physicochemical filters:

• Molecular weight: Range specification (Da)
• LogP: Partition coefficient limits
• TPSA: Polar surface area constraints
• HBD/HBA: Hydrogen bond donor/acceptor counts
• Rotatable bonds: Molecular flexibility limits

Structural filters:

• Ring count: Number of rings in molecule
• Heavy atoms: Total non-hydrogen atoms
• Formal charge: Ionic species filtering
• Isotope content: Labeled compound exclusion

Combined search approaches

Multi-criteria filtering:

1. Start with broad text/name search
2. Apply property ranges to focus results
3. Use structural filters for specific chemotypes
4. Sort by relevance or specific properties

Iterative refinement:

1. Initial broad search to assess result scope
2. Progressive filter application
3. Result quality assessment
4. Query modification based on findings

Batch search capabilities

Multiple compound processing

A compound lookup takes up to 10 names, SMILES or CIDs at once ("Look up aspirin, ibuprofen and caffeine in PubChem"), queries PubChem for them in parallel and merges the results under one result card. For more than 10 compounds, or for a list pasted or uploaded in chat, ask Ivy to run the lookups in batches of 10.

Batch result management

Result aggregation:

• Consolidated property tables across all compounds
• Missing data handling and reporting
• Statistical summaries of property distributions

Export preparation:

• Filtered result subsets
• Property selection for downstream tools
• Format conversion (SMILES, SDF, CSV)

Search optimization techniques

Performance considerations

Query efficiency:

• Use specific identifiers (CID, CAS) over text searches
• Limit result sets with property filters early
• Avoid overly broad queries that return thousands of results

Caching strategies:

• .bloom* caches PubChem results to improve performance
• Recent searches load faster on repeat
• Large batch jobs benefit from result persistence

Result quality assessment

Data completeness checking:

• Review available properties for each compound
• Identify compounds with missing critical data
• Cross-reference with alternative data sources

Structure validation:

• Compare returned structures with expected molecules
• Check for stereochemistry preservation
• Validate identifier-structure consistency

Specialized search patterns

Drug discovery applications

Lead compound verification:

1. Search by compound name or CAS number
2. Verify structure and properties match expectations
3. Check for bioassay data availability
4. Export standardized identifiers for further analysis

Analog series identification:

1. Search for core scaffold or pharmacophore
2. Apply property filters matching lead compound profile
3. Review structural similarity and diversity
4. Export hit list for virtual screening

Library curation workflows

Compound collection enrichment:

1. Batch search for compound identifiers
2. Add standardized properties and identifiers
3. Fill missing data gaps with calculated values
4. Generate quality control reports

Vendor catalog validation:

1. Search supplier-provided identifiers
2. Cross-reference with PubChem data
3. Identify discrepancies or data quality issues
4. Standardize compound representations

Troubleshooting search issues

Common problems and solutions

No results found:

• Check identifier format and validity
• Try alternative identifiers (name vs CAS vs CID)
• Remove special characters or formatting issues
• Consider if compound is in PubChem database

Too many results:

• Add property filters to narrow scope
• Use more specific search terms
• Switch from text to identifier-based search
• Apply structural constraints

Incorrect structures:

• Verify SMILES/InChI syntax
• Check stereochemistry specifications
• Compare with known reference structures
• Contact support for database correction requests

Performance issues:

• Reduce batch size for large searches
• Use caching for repeated queries
• Implement result pagination
• Schedule large jobs during off-peak hours

Integration with .bloom* workflows

Seamless data flow

Compound management:

• Direct import to .bloom*'s compound database
• Integration with screening result analysis
• Connection to property prediction tools

Research pipeline:

• Feed search results to ADMET prediction
• Integration with virtual screening campaigns
• Connection to procurement and ordering systems

Data standardization:

• Consistent identifier formats across projects
• Reliable property calculations
• Quality-controlled compound information