Using Statistical Averages to Improve Similarity Search of Large DNA Molecules to the Human Genome

• Activity Number: 244 - Statistical methods for microbiome data analysis and beyond
• Type: Contributed
• Date/Time: Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Albertas Dvirnas*
• Companies: Lund University
• Keywords: optical DNA mapping; similarity search; cross correlation; time series; averaging; competitive binding

Abstract

Optical DNA mapping (ODM) allows visualization of long-range sequence information along single DNA molecules. The data can for example be used for detecting long range structural variations, for aiding DNA sequence assembly of complex genomes and for mapping epigenetic marks and DNA damage across the genome. Here we use a competitive binding approach, based on YOYO-1 and netropsin, which highlights the contour of the DNA molecules, while simultaneously creating a continuous sequence specific pattern, based on the AT/GC variation along the detected molecule. In this work we discuss how statistical averaging of optical maps coming from many different molecules can be used to improve the underlying theory model and improve the overall accuracy of the similarity search ( Pearson Cross correlation, Dynamical Time Warping and Matrix Profile) of the single DNA molecules to the human genome.