Probability of coding of a DNA sequence: an algorithm to predict translated reading frames from their thermodynamic...

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Identifying protein-coding regions in DNA is crucial for understanding gene function.
• Existing methods often rely on sequence periodicity, which can be confounded by overlapping or complementary frames.
• Thermodynamic characteristics of translated reading frames offer a potential alternative for accurate prediction.

Purpose of the Study:

• To develop and present an algorithm for calculating the probability of a DNA reading frame encoding a protein.
• To establish a prediction procedure for differentiating coding from non-coding reading frames.
• To ensure the prediction method is robust against biases from overlapping or complementary DNA strands.

Main Methods:

• Algorithm development based on thermodynamic characteristics of translated reading frames.
• Implementation of a prediction procedure focusing on the putative protein product's features.
• Validation against known coding and non-coding sequences, considering potential biases.

Main Results:

• An algorithm accurately estimates the probability of a reading frame being protein-coding.
• A prediction procedure successfully distinguishes coding from non-coding frames.
• The method's independence from sequence periodicity avoids biases from overlapping or complementary frames.

Conclusions:

• The presented algorithm and prediction procedure offer a reliable approach for identifying protein-coding DNA sequences.
• Analyzing thermodynamic properties provides a powerful, bias-free method for gene prediction.
• This tool enhances genomic analysis by accurately identifying functional reading frames.