INTEGRATING COMPUTATIONAL PROTEIN FUNCTION PREDICTION INTO DRUG DISCOVERY INITIATIVES.

Area of Science:

• Bioinformatics
• Computational Biology
• Drug Discovery

Background:

• Genomic sequencing and structural genomics generate vast protein data.
• Traditional homology-based protein function annotation faces limitations with increasing sequence diversity.
• Accurate protein function prediction is crucial for identifying drug targets.

Purpose of the Study:

• To review current bioinformatics programs and approaches for protein function prediction and annotation.
• To discuss the integration of these methods into drug discovery initiatives.
• To highlight the importance of non-homology based methods.

Main Methods:

• Review of existing literature on bioinformatics tools for protein function prediction.
• Analysis of non-homology based methods utilizing sequence features, structure, evolution, and biochemical/genetic knowledge.
• Discussion of the application of these methods in drug discovery pipelines.

Main Results:

• Homology-based methods are insufficient due to expanding sequence databases.
• Emergence of diverse non-homology based prediction methods.
• Bioinformatic tools are increasingly vital for target identification and lead discovery.

Conclusions:

• Computational protein function prediction is essential for modern drug discovery.
• Integrating diverse bioinformatics approaches enhances target identification efficiency.
• Advanced annotation of protein functional sites is key to leveraging genomic data.