Optimized topographical labeling for identification of unknown nucleic acids

Abstract

Detection of nucleic acid biomarker has particular importance for disease diagnosis and environmental safety. Yet the identification of low-abundant nucleic acids with unknown sequence remains extremely challenging. Previously we reported a strategy to identify short cDNA molecules using nicking endonucleases that create sequence-specific topographical labels visible by AFM. As such, the identity of the unknown cDNA molecules can be determined based on the labeling patterns composed by the locations of and the spacing between the labels. Here, in order to improve the accuracy, we optimized all possible combinations from total 8 types of nicking endonucleases. The results showed that, based on an optimized combination, NtBsmAI + NtBstNBI + NtAlwI, the average accuracy was improved to 97.1 ± 5.6% as compared with the average accuracy of 51.2 ± 34.4% where only one type of nicking endonuclease was used. More importantly, this improved accuracy is applicable to more than over 85% of human mRNA transcripts. Together, our study suggests an optimization strategy for detecting nucleic acid with unknown sequence using AFM-based measurement, with applications for identifying pathogen and disease at single molecule level.