ThermoPred

ThermoPred predicts thermophilic proteins from amino acid sequence information to identify thermostable enzymes for protein engineering.

Key Features:

• Sequence-based prediction: Classifies proteins as thermophilic or non-thermophilic using amino acid sequence information.
• Support vector machine (SVM): Employs an SVM classification algorithm for prediction.
• Amino acid distribution features: Uses amino acid distribution data as input features.
• Selected amino acid pair features: Incorporates selected amino acid pairs as input features.
• Benchmark dataset: Trained and tested on a dataset comprising 915 thermophilic proteins and 793 non-thermophilic proteins.
• Validation and performance: Evaluated by jackknife cross-validation with 93.8% correct prediction rate for thermophilic proteins and 92.7% for non-thermophilic proteins.

Scientific Applications:

• Enzyme engineering: Identifies thermostable enzymes for enzyme engineering efforts.
• Protein design: Aids protein design by identifying sequence determinants associated with thermostability.
• Mesophilic-to-thermostable engineering: Identifies mesophilic proteins that could be engineered into thermostable variants.
• Industrial biotechnology and synthetic biology: Applicable to development of enzymes with improved thermal resistance for industrial biotechnology and synthetic biology.

Methodology:

Support vector machine (SVM) classification using features derived from amino acid distribution and selected amino acid pairs, trained and tested on a benchmark dataset of 915 thermophilic and 793 non-thermophilic proteins and evaluated by jackknife cross-validation.