BLAST quickstart

You want to find sequences similar to your query and know what they are — BLAST does it in a few steps: enter a sequence or accession, pick a program and database (blastp against SwissProt covers most protein work), run the search, and explore the ranked hits with their scores and alignments.

Webapp input methods

The .bloom* webapp provides a comprehensive interface for BLAST sequence homology searches.

Input type selection

From previous job

• Job Output Selector: Browse completed jobs with protein sequence outputs
• Data type: Requires protein_sequence or fasta output
• Compatible tools: ESMFold, other protein analysis jobs

Webapp parameters

Basic parameters

Program descriptions

Database descriptions

Job options

• Job Title (optional): Custom identifier for tracking
• Advanced settings…: Opens settings panel for additional BLAST parameters
• Use Personalized Settings: Toggle for project-specific configuration

Personalized settings

When enabled, loads user-specific BLAST settings from Supabase including:

• Custom scoring matrices
• Composition-based statistics options
• Gap costs and word size preferences

Submit button

• Label: "Run BLAST"
• Disabled when: Manual input empty OR no job output selected

Sequence similarity searching with BLAST in .bloom* platform

Step-by-step sequence analysis workflow

Common BLAST search scenarios

Protein function identification

1. Input novel protein sequence from genomics
2. Search against UniProt/SwissProt database
3. Identify top hits with known functions
4. Analyze conserved domains and motifs
5. Generate functional annotation report

Homolog discovery

1. Upload protein of interest
2. Search against organism-specific databases
3. Filter by identity and coverage thresholds
4. Identify paralogs and orthologs
5. Analyze evolutionary relationships

Contamination detection

1. Input sequencing reads or assembled contigs
2. Search against common contaminant databases
3. Identify potential contamination sources
4. Filter and clean sequence datasets
5. Generate contamination reports

Primer design validation

1. Input primer sequences
2. Search against target genome databases
3. Check specificity and potential off-target binding
4. Optimize primer design based on results
5. Validate primer specificity

Quality control and validation

Query preparation checklist

Sequence quality:

• Valid FASTA format with proper headers
• Correct sequence alphabet (amino acids or nucleotides)
• Appropriate sequence length for search type
• No ambiguous characters or gaps

Search parameters:

• Appropriate database selection for query type
• Reasonable E-value threshold for sensitivity
• Suitable word size and gap penalties
• Proper taxonomic or compositional filtering

Result validation

Hit quality assessment:

• Review top hit E-values and bit scores
• Check sequence identity and coverage percentages
• Validate alignment quality and gap patterns
• Assess biological relevance of top hits

Statistical validation:

• Compare scores across different database searches
• Check for consistent hit patterns
• Validate against known positive controls
• Assess statistical significance of findings

Advanced search strategies

Multi-database searching

• Search across multiple databases simultaneously
• Combine results from different sequence collections
• Cross-validate hits across database types
• Generate comprehensive similarity profiles

Iterative refinement

• Use top hits as queries for follow-up searches
• Refine search parameters based on initial results
• Expand or restrict taxonomic scope
• Adjust sensitivity for specific discovery goals

Comparative analysis

• Search same query against different databases
• Compare hit distributions across organisms
• Identify database-specific vs. universal hits
• Generate comparative similarity reports

Integration with .bloom* workflows

Upstream sequence sources

Genomics data:

• Assembled contigs: Genome assembly outputs
• Transcript sequences: RNA-seq derived sequences
• Protein predictions: Gene finding and translation results
• Metagenomic data: Environmental sequencing samples

Protein data:

• Novel proteins: Discovery proteomics results
• Predicted structures: ESMFold sequence inputs
• Mutant libraries: Engineered sequence variants
• Synthetic designs: De novo protein sequences

Downstream applications

Functional annotation:

• Gene ontology mapping: Link sequences to biological functions
• Pathway analysis: Identify metabolic and signaling pathways
• Domain identification: Find conserved structural elements
• Motif discovery: Identify regulatory sequence patterns

Comparative analysis:

• Phylogenetic studies: Build evolutionary relationships
• Sequence clustering: Group similar sequences
• Conservation analysis: Identify functionally important regions
• Species distribution: Analyze taxonomic distribution patterns

Structure prediction:

• Homology modeling: Use templates from BLAST hits
• Domain prediction: Identify structural domain boundaries
• Secondary structure: Predict folding patterns
• Functional site identification: Locate active sites and interfaces

Performance optimization

Search efficiency

Database selection:

• Choose appropriate database size for query scope
• Use taxonomic restrictions to limit search space
• Select program type matching query-database types
• Consider pre-filtered databases for specific applications

Parameter optimization:

• Adjust E-value thresholds for desired sensitivity
• Set appropriate max target sequences
• Use composition-based statistics for biased sequences
• Enable filtering for repetitive or low-complexity regions

Result processing

Batch processing:

• Process multiple queries simultaneously
• Use result filtering to manage output volume
• Implement result caching for repeated searches
• Automate downstream analysis pipelines

Output management:

• Configure appropriate output formats
• Use tabular formats for programmatic analysis
• Generate summary statistics automatically
• Implement result filtering and prioritization

Best practices for sequence similarity searching

Query preparation

Sequence optimization:

• Use high-quality, full-length sequences when possible
• Remove vector sequences and adapters
• Mask repetitive or low-complexity regions
• Consider domain-specific searches for multi-domain proteins

Parameter selection:

• Start with default parameters for initial searches
• Adjust sensitivity based on sequence conservation
• Use appropriate substitution matrices
• Consider query length and composition

Result interpretation

Biological context:

• Interpret results in context of known biology
• Cross-reference with literature and databases
• Consider experimental validation approaches
• Assess functional relevance of hits

Statistical considerations:

• Understand E-value and bit score meanings
• Consider multiple testing corrections
• Account for database composition effects
• Validate against known reference sequences

Research reproducibility

Documentation:

• Record all search parameters and settings
• Preserve input sequences and database versions
• Document result interpretation criteria
• Create analysis notebooks for methods

Collaboration:

• Share search strategies and parameters
• Generate standardized result reports
• Document hit validation approaches
• Maintain searchable result archives

Troubleshooting common issues

No significant hits

• Lower E-value threshold for more sensitivity
• Check sequence quality and remove contaminants
• Try different substitution matrices
• Search against more comprehensive databases

Too many hits

• Raise E-value threshold for specificity
• Use taxonomic restrictions
• Apply compositional filtering
• Consider domain-specific searches

Poor alignment quality

• Check sequence quality and correct errors
• Use appropriate gap penalties
• Try different alignment parameters
• Consider if sequences are too divergent

Performance issues

• Use smaller databases for initial searches
• Implement result limits
• Consider iterative search strategies
• Use pre-filtered databases when appropriate

Read next