Resolution-Free Accurate DNA Contour Length Estimation from Atomic Force Microscopy Images.

摘要: This research presented an accurate and efficient contour length estimation method developed for DNA digital curves acquired from Atomic Force Microscopy (AFM) images. automation is calibrated against different AFM resolutions ideal to be extended all kinds of biopolymer samples, encompassing sample stiffnesses. The methodology considers the curve local geometric relationship, as these shape segments pixel connections represent actual morphology it being imaged scanning. In order incorporate true geometry relationship that embedded in continuous form original sample, one needs find this counterpart digitized image. realized by taking skeleton backbone using pixels’ connection its representation. research, uses 8-connect Freeman Chain Code (CC) describe directional between image pixels, account shapes four connected pixels. result a novel number (SN) system derived CC, which fully automated algorithm can applied samples any estimation, with computational cost. shape-wise consideration weighted modify great precision, accounting morphologies resulted error falls below 0.07%, magnitude improvement compared previous findings.