Motif caller for sequence reconstruction in motif-based DNA storage

Area of Science:

• Biotechnology
• Bioinformatics
• Data Science

Background:

• DNA data storage offers exceptional durability for long-term archiving.
• Current DNA data storage methods face bottlenecks in synthesis speed and cost.
• Existing data retrieval relies on basecalling, which is inefficient for motif-based DNA.

Purpose of the Study:

• To develop a more accurate and efficient method for reading data from motif-based DNA storage.
• To bypass the limitations of traditional basecalling for motif detection.

Main Methods:

• Introduced Motif Caller, a machine learning model.
• Designed Motif Caller to directly detect entire DNA motifs from raw nanopore sequencing signals.
• Bypassed the intermediate step of basecalling individual DNA bases.

Main Results:

• Motif Caller significantly improves accuracy by leveraging richer signal features.
• The direct motif detection approach enhances data retrieval efficiency.
• Eliminates the imprecision and inefficiency of two-step, motif-agnostic basecalling.

Conclusions:

• Motif Caller represents a significant advancement in DNA data storage retrieval.
• Direct motif detection offers a more precise and efficient solution for motif-based DNA storage systems.
• This method addresses key bottlenecks in current DNA data archiving technologies.