Cell Lines CoCoPUTs: A Database of Codon and Codon-pair Usage Frequencies in Cell Lines.

Area of Science:

• Genomics
• Bioinformatics
• Cell Biology

Background:

• Cell lines are crucial for biological research, drug discovery, and biologics production.
• Understanding cell line characteristics aids in data interpretation and experimental design.

Purpose of the Study:

• To introduce Cell Lines CoCoPUTs, a comprehensive database of transcriptomic-weighted codon and codon pair usage for 1,866 cell lines.
• To analyze codon usage patterns across different cell line databases and investigate clustering based on origin and disease phenotype.

Main Methods:

• Compiled codon and codon pair usage data for 1,866 cell lines from COSMIC, CCLE, and HPA databases.
• Applied unsupervised machine learning (hierarchical and spectral clustering) to analyze codon usage in 1,355 non-metastatic cell lines.
• Performed distance-based comparisons of codon usage versus codon pair usage.

Main Results:

• Broadly similar codon usage distributions were observed across overlapping cell lines from different databases, despite variations in data sources and analysis platforms (microarray vs. RNA-SEQ).
• Unsupervised clustering revealed more distinct groupings based on codon pair usage compared to codon usage for non-metastatic cell lines.
• Distance-based analysis indicated that codon usage provides comparable or smaller within-group distances than codon pair usage.
• Cell lines demonstrated a closer relationship to their tissue of origin than to their disease phenotype.

Conclusions:

• Cell Lines CoCoPUTs provides a valuable resource for analyzing codon usage in diverse cell lines.
• Codon usage patterns can effectively distinguish cell lines, with origin being a stronger determinant than disease phenotype.
• The findings highlight the utility of codon usage analysis in cell line characterization and comparative genomics.