construct-generator

name: construct-generator description: Generate expected plasmid maps by combining vector backbones with insert sequences. Use when the user asks about in-silico cloning, simulating cloning reactions, generating construct maps, or verifying clone sequences against expected products. allowed-tools: Bash(construct_generator:*)

Construct Generator

Generate expected plasmid maps by combining vector backbones with insert sequences. Essential for clone verification workflows.

CLI Location

./construct_generator [flags]

What It Does

The construct generator creates in-silico plasmid maps representing the expected product of a cloning reaction. It:

1. Loads a vector backbone (from built-in library or local files)
2. Fetches an insert sequence (from NCBI or local file)
3. Simulates the cloning reaction based on method (LIC, Gibson, SLIC, restriction)
4. Outputs a GenBank file with proper annotations

Usage

List Available Vectors

./construct_generator -list-vectors

Shows all vectors from:

• Built-in defaults (438 series, common expression vectors)
• ~/.benchaid/vectors.yaml (user-defined)
• vectors/*.yaml (project-specific)

Generate a Construct

From NCBI accession:

./construct_generator \
  -vector 438-A \
  -insert-accession NM_007192 \
  -output my_construct.gb

From local file:

./construct_generator \
  -vector 438-A \
  -insert-file insert.fasta \
  -gene-name MyProtein \
  -output my_construct.gb

Cloning Methods

Key Flags

Vector Library

Built-in 438 Series

The 438 series vectors are bacterial expression vectors with various fusion tags:

Local Vectors

GenBank files in ./vectors/ are automatically available:

• vectors/438-A.gb through vectors/438-H.gb

Adding Custom Vectors

Create a YAML file in vectors/ or ~/.benchaid/vectors.yaml:

vectors:
  my-vector:
    file: path/to/vector.gb
    insert_site: 1234  # Position for insert
    method: lic        # Default cloning method
    description: "My custom expression vector"

Validation

When -validate is enabled, the tool checks:

1. Frame check: Insert is in-frame with vector ORF
2. Internal stops: No premature stop codons in the fusion
3. Insert length: Reasonable size for expression

Output Format

The output GenBank file includes:

• Full plasmid sequence with insert
• Annotated features:
  • Vector backbone elements
  • Insert CDS with gene name
  • Fusion tags
  • Promoters, terminators, origins
• Metadata in COMMENT section

Example Workflows

Standard Cloning Project

# 1. Design primers for the insert
./primer_cli.py --accession NM_007192 --index 1

# 2. Generate expected construct
./construct_generator \
  -vector 438-B \
  -insert-accession NM_007192 \
  -output expected_construct.gb

# 3. After cloning and sequencing, verify with orf_verifier
python3 scripts/orf_verifier_cli.py \
  --plasmid expected_construct.gb \
  --protein MKHHHHHHMS...

Using Local Insert File

# Generate construct from synthesized gene
./construct_generator \
  -vector 438-A \
  -insert-file synthesized_gene.fasta \
  -gene-name DSIF \
  -method lic \
  -output dsif_construct.gb

Batch Processing

# Generate constructs for multiple accessions
for acc in NM_007192 NM_001234 NM_005678; do
  ./construct_generator \
    -vector 438-B \
    -insert-accession $acc \
    -output constructs/${acc}.gb
done

Integration with Other Tools

Troubleshooting

"Vector not found"

• Run -list-vectors to see available vectors
• Check that vector files exist in vectors/ directory
• Ensure GenBank files have proper LOCUS line

"Could not fetch accession"

• Verify the accession exists at NCBI
• Check internet connectivity
• Use -insert-file with a local file instead

"Frame validation failed"

• The insert may not be in-frame with vector
• Check primer design includes correct reading frame
• Use -validate=false to skip validation if intentional