Hipsec

Hipsec predicts the potential for high-level production of extracellular proteins via gene overexpression in Aspergillus niger using protein sequence-derived features.

Key Features:

• Sequence-based prediction: Uses machine learning on protein sequences to forecast high-level extracellular protein production in Aspergillus niger based on amino acid composition and sequence-derived features.
• FASTA input format: Accepts protein sequences provided in FASTA format for analysis.
• Training dataset: Trained and evaluated on experimental overexpression data including over 600 homologous and nearly 2,000 heterologous fungal genes expressed in Aspergillus niger using a standardized expression cassette.
• Feature identification: Identifies DNA and protein sequence features that correlate with successful or unsuccessful production outcomes.
• Observed residue correlations: Reports positive correlations of tyrosine and asparagine compositions with high-level production and negative correlations of methionine and lysine with unsuccessful production.

Scientific Applications:

• Predicting expression outcomes: Predicts whether overexpression of a given gene will yield high-level extracellular protein production in Aspergillus niger.
• Protein engineering guidance: Informs sequence-based protein engineering strategies to alter amino acid composition toward features associated with higher production.
• Industrial enzyme optimization: Supports selection and design of enzymes for improved production levels in industrial bioprocesses using Aspergillus niger expression systems.

Methodology:

Applies sequence-based machine learning on amino acid composition and DNA/protein sequence features derived from experimental overexpression of over 600 homologous and nearly 2,000 heterologous fungal genes in Aspergillus niger using a standardized expression cassette.