SPiCE

SPiCE provides sequence-based protein classification and feature extraction to enable predictive modeling of protein properties from amino acid sequences.

Key Features:

• Integrated environment: Integrates data upload/download, sequence feature calculation, feature visualization, classifier construction, and classifier testing.
• Feature calculation and visualization: Calculates sequence-based features, including amino acid composition, and visualizes these features within predefined classes to explore property differences.
• Classifier construction and testing: Constructs and tests classifiers using derived features to predict protein properties such as subcellular localization and solubility.

Scientific Applications:

• Exploratory sequence-feature analysis: Explores sequence-based features to identify patterns associated with different biological outcomes.
• Production-level prediction in fungi: Distinguishes proteins associated with low versus high production levels in fungi.
• Expression-level analysis in yeast: Differentiates proteins with varying expression levels in yeast.
• Property prediction: Predicts protein properties such as subcellular localization and solubility from sequence-derived features.

Methodology:

Calculates sequence-based features from amino acid sequences and derived features, visualizes features per class, and applies classifier algorithms to construct and test predictive models of protein properties.