Analysis on the distribution of bases in 1487 human protein coding sequences

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Human protein-coding sequences (CDS) are fundamental to genetic function.
• Understanding base composition (adenine, cytosine, guanine, thymine) is crucial for genomic analysis.
• Previous studies have explored base frequencies, but a comprehensive spatial analysis is less common.

Purpose of the Study:

• To analyze the occurrence frequencies of bases A, C, G, and T in 1487 human protein-coding sequences.
• To assess the distribution of these base frequencies using a novel 3-D diagrammatic method.
• To identify any deviations from established empirical rules regarding base composition.

Main Methods:

• Calculation and analysis of base occurrence frequencies (a, c, g, t) for 1487 human CDS.
• Representation of each CDS as a point in 3-D space for visualization of base frequency distribution.
• Determination of radical distance and analysis of radical distribution in 3-D space.
• Application of an empirical rule (a² + c² + g² + t² < 1/3) to the dataset.

Main Results:

• The radical distribution of base frequencies for the 1487 human CDS points is normal, centered around the origin (a=c=g=t=1/4).
• An empirical rule, a² + c² + g² + t² < 1/3, was found to hold true for 1486 out of 1487 sequences.
• The human parathymosin protein coding sequence was the sole exception, failing to adhere to this empirical rule.
• Preliminary analysis of amino acid composition and structural class for parathymosin was conducted.

Conclusions:

• Human protein-coding sequences generally follow predictable base frequency patterns visualized in 3-D space.
• The identified empirical rule provides a useful metric for assessing base composition normality in CDS.
• The human parathymosin sequence represents a unique case warranting further investigation into its distinct base composition and potential functional implications.