SUP: a probabilistic framework to propagate genome sequence uncertainty, with applications.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Genetic sequencing, especially next-generation methods, introduces errors and uncertainties in base calls despite high read counts.
• Current genetic analyses often overlook these inherent uncertainties, assuming sequences are error-free.
• Imperfect sequencing coverage leads to unreliable base calls, affecting the accuracy of downstream genetic studies.

Purpose of the Study:

• To demonstrate how sequencing uncertainty affects downstream genetic analyses.
• To propose a straightforward method for propagating sequencing uncertainty in genetic data.
• To provide a more accurate evaluation of errors in genetic analyses.

Main Methods:

• Developed Sequence Uncertainty Propagation (SUP), a method using probabilistic matrix representation of sequences.
• Incorporated base quality scores into the matrix to quantify uncertainty.
• Utilized resampling and replication as a framework for uncertainty propagation, mimicking bootstrapping.

Main Results:

• Demonstrated the impact of sequencing uncertainty on SARS-CoV-2 data analysis.
• Showed that SUP adds only a linear computational cost.
• Revealed that SARS-CoV-2 lineage designations and clock rate estimates are more variable than previously reported.

Conclusions:

• Ignoring sequencing uncertainty can lead to overly confident and inaccurate conclusions in genetic analyses.
• SUP provides a robust framework for incorporating base call uncertainty, enhancing the reliability of genetic findings.
• The method is crucial for accurate phylogenetic and epidemiological studies, particularly for rapidly evolving viruses like SARS-CoV-2.