Invention:
This invention is for the improvement of nanopore sequencing base calling accuracy via an iterative approach involving the use of existing machine learning algorithms and alignment with ground-truth references. This new method allows base calling accuracy to improve as the technology is trained, which in turn creates countless opportunities for sequencing improvement in a variety of applications.
Background:
The concept of nanopore sequencing was first described in publication by Branton, Deamer et al in PNAS in 1996. The research group showed translocation of nucleic acids through a nanopore set in a lipid bilayer, and noted that "Channel blockades can therefore be used to measure polynucleotide length. In 1998 US Patent 5,795,782, "Characterization of Individual Polymer Molecules Based on Monomer-Interface Interactions" was granted as one of the first nanopore sequencing technology patents. Since this discovery, the accuracy of base calling has decreased due to chemically modified nucleotides which have caused significant restriction of DNA/RNA investigation with nanopore sequencing. This novel method polishes yielded sequences, comparing them to references that allows for accuracy convergence over time.
Applications:
- Training of nanopore sequencing base calling software via machine learning
Advantages:
- System is trained over time improving accuracy
- Improved sequencing of DNA/RNA
- Utilizes common exiting ML technology
- Allows for determination of therapeutic RNA agents quality
