Image analysis and length estimation of biomolecules using AFM

Advancing High-Throughput Cellular Atomic Force Microscopy with Automation and Artificial Intelligence

Atomic force microscopy (AFM) has reached a significant level of maturity in biology, demonstrated by the diversity of modes for obtaining not only topographical images but also insightful mechanical and adhesion data by performing force measurements on delicate samples with a controlled environment (e.g., liquid, temperature, pH). Numerous studies have applied AFM to describe biological phenomena at the molecular and cellular scales, and even on tissues. Despite these advances, AFM is not established as a diagnostic tool in the biomedical field. This article describes the reasons for this gap, focusing on one of the main weaknesses of bio-AFM: its low data throughput. We review current efforts to improve the automation of AFM measurements in particular on living cells, as well as the developments in automating data analysis. For the latter, artificial intelligence (AI) is progressively employed to classify data to distinguish healthy and diseased cells or tissues. Finally, we propose a roadmap to foster the application of bio-AFM into medical diagnostics.

AI-based atomic force microscopy image analysis allows to predict electrochemical impedance spectra of defects in tethered bilayer membranes

Atomic force microscopy (AFM) image analysis of supported bilayers, such as tethered bilayer membranes (tBLMs) can reveal the nature of the membrane damage by pore-forming proteins and predict the electrochemical impedance spectroscopy (EIS) response of such objects. However, automated analysis involving pore detection in such images is often non-trivial and can require AI-based object detection techniques. The specific object-detection algorithm we used to determine the defect coordinates in real AFM images was a convolutional neural network (CNN). Defect coordinates allow to predict the EIS response of tBLMs populated by the pore-forming toxins using finite element analysis (FEA) modeling. We tested if the accuracy of the CNN algorithm affected the EIS spectral features sensitive to defect densities and other physical parameters of tBLMs. We found that the EIS spectra can be predicted sufficiently well, however, systematic errors of characteristic spectral points were observed and need to be taken into account. Importantly, the comparison of predicted EIS curves with experimental ones allowed to estimate important physical parameters of tBLMs such as the specific resistance of submembrane reservoir. This reservoir separates phospholipid bilayer from the solid support. We found that the specific resistance of the reservoir amounts to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$10^{4.25 \pm 0.10}$$\end{document}104.25±0.10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Omega \cdot cm$$\end{document}Ω·cm which is approximately two orders of a magnitude higher compared to the specific resistance of the buffer bathing tBLMs studied in this work. We hypothesize that such effect may be related in part due to decreased concentration of ionic carriers in the submembrane due to decreased relative dielectric permittivity in this region.

Supervised Learning for Predictive Pore Size Classification of Regenerated Cellulose Membranes Based on Atomic Force Microscopy Measurements

Nanoporous dialysis membranes made of regenerated cellulose are used as molecular weight cutoff standards in bioseparations. In this study, mesoporous standards with Stokes' radii (50 kDa/2.7 nm, 100 kDa/3.4 nm and 1000 kDa/7.3 nm) and overlapping skewed distributions were characterized using AFM, with the specific aim of generating pore size classifiers for biomimetic membranes using supervised learning. Gamma transformation was used prior to conducting discriminant analysis in terms of the area under the receiver operating curve (AUC) and classification accuracy (Acc). Monte Carlo simulations were run to generate datasets (n = 10) on which logistic regression was conducted using a constant ratio of 80:20 (measurement:algorithm training), followed by algorithm validation by WEKA. The proposed algorithm can classify the 1000 kDa vs. 100 kDa (AUC > 0.8) correctly, but discrimination is weak for the 100 kDa vs. 50 kDa (AUC < 0.7), the latter being attributed to the instrument accuracy errors below 5 nm. As indicated by the results of the cross-validation study, a test size equivalent to 70% (AUC_tapping = 0.8341 ± 0.0519, Acc_tapping = 76.8% ± 5.9%) and 80% (AUC_fluid = 0.7614 ± 0.0314, Acc_tfluid = 76.2% ± 1.0%) of the training sets for the tapping and fluid modes are needed for correct classification, resulting in predicted reduction of scan times.

Digital Polymerase Chain Reaction Paired with High-Speed Atomic Force Microscopy for Quantitation and Length Analysis of DNA Length Polymorphisms

DNA length polymorphisms are found in many serious diseases, and assessment of their length and abundance is often critical for accurate diagnosis. However, measuring their length and frequency in a mostly wild-type background, as occurs in many situations, remains challenging due to their variable and repetitive nature. To overcome these hurdles, we combined two powerful techniques, digital polymerase chain reaction (dPCR) and high-speed atomic force microscopy (HSAFM), to create a simple, rapid, and flexible method for quantifying both the size and proportion of DNA length polymorphisms. In our approach, individual amplicons from each dPCR partition are imaged and sized directly. We focused on internal tandem duplications (ITDs) located within the FLT3 gene, which are associated with acute myeloid leukemia and often indicative of a poor prognosis. In an analysis of over 1.5 million HSAFM-imaged amplicons from cell line and clinical samples containing FLT3-ITDs, dPCR-HSAFM returned the expected variant length and variant allele frequency, down to 5% variant samples. As a high-throughput method with single-molecule resolution, dPCR-HSAFM thus represents an advance in HSAFM analysis and a powerful tool for the diagnosis of length polymorphisms.

DNA accessibility of chromatosomes quantified by automated image analysis of AFM data

DNA compaction and accessibility in eukaryotes are governed by nucleosomes and orchestrated through interactions between DNA and DNA-binding proteins. Using QuantAFM, a method for automated image analysis of atomic force microscopy (AFM) data, we performed a detailed statistical analysis of structural properties of mono-nucleosomes. QuantAFM allows fast analysis of AFM images, including image preprocessing, object segmentation, and quantification of different structural parameters to assess DNA accessibility of nucleosomes. A comparison of nucleosomes reconstituted with and without linker histone H1 quantified H1's already described ability of compacting the nucleosome. We further employed nucleosomes bearing two charge-modifying mutations at position R81 and R88 in histone H2A (H2A R81E/R88E) to characterize DNA accessibility under destabilizing conditions. Upon H2A mutation, even in presence of H1, the DNA opening angle at the entry/exit site was increased and the DNA wrapping length around the histone core was reduced. Interestingly, a distinct opening of the less bendable DNA side was observed upon H2A mutation, indicating an enhancement of the intrinsic asymmetry of the Widom-601 nucleosomes. This study validates AFM as a technique to investigate structural parameters of nucleosomes and highlights how the DNA sequence, together with nucleosome modifications, can influence the DNA accessibility.

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

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

Length computation of irradiated plasmid DNA molecules

Compromised detection of short DNA fragments can result in underestimation of radiation-induced clustered DNA damage. The fragments can be detected with atomic force microscopy (AFM), followed by image analysis to compute the length of plasmid molecules. Plasmid molecules imaged with AFM are represented by open or closed curves, possibly with crossings. For the analysis of such objects, a dedicated algorithm was developed, and its usability was demonstrated on the AFM images of plasmid pBR322 irradiated with ⁶⁰Co gamma rays. The analysis of the set of the acquired AFM images revealed the presence of DNA fragments with lengths shorter than 300 base pairs that would have been neglected by a conventional detection method.

DNA nanomapping using CRISPR-Cas9 as a programmable nanoparticle

Progress in whole-genome sequencing using short-read (e.g., <150 bp), next-generation sequencing technologies has reinvigorated interest in high-resolution physical mapping to fill technical gaps that are not well addressed by sequencing. Here, we report two technical advances in DNA nanotechnology and single-molecule genomics: (1) we describe a labeling technique (CRISPR-Cas9 nanoparticles) for high-speed AFM-based physical mapping of DNA and (2) the first successful demonstration of using DVD optics to image DNA molecules with high-speed AFM. As a proof of principle, we used this new “nanomapping” method to detect and map precisely BCL2–IGH translocations present in lymph node biopsies of follicular lymphoma patents. This HS-AFM “nanomapping” technique can be complementary to both sequencing and other physical mapping approaches.

Physical mapping of DNA can be used to detect structural variants and for whole-genome haplotype assembly. Here, the authors use CRISPR-Cas9 and high-speed atomic force microscopy to ‘nanomap’ single molecules of DNA.

Algorithmic methods to infer the evolutionary trajectories in cancer progression

The genomic evolution inherent to cancer relates directly to a renewed focus on the voluminous next-generation sequencing data and machine learning for the inference of explanatory models of how the (epi)genomic events are choreographed in cancer initiation and development. However, despite the increasing availability of multiple additional -omics data, this quest has been frustrated by various theoretical and technical hurdles, mostly stemming from the dramatic heterogeneity of the disease. In this paper, we build on our recent work on the "selective advantage" relation among driver mutations in cancer progression and investigate its applicability to the modeling problem at the population level. Here, we introduce PiCnIc (Pipeline for Cancer Inference), a versatile, modular, and customizable pipeline to extract ensemble-level progression models from cross-sectional sequenced cancer genomes. The pipeline has many translational implications because it combines state-of-the-art techniques for sample stratification, driver selection, identification of fitness-equivalent exclusive alterations, and progression model inference. We demonstrate PiCnIc's ability to reproduce much of the current knowledge on colorectal cancer progression as well as to suggest novel experimentally verifiable hypotheses.

Histological Image Processing Features Induce a Quantitative Characterization of Chronic Tumor Hypoxia

Hypoxia in tumors signifies resistance to therapy. Despite a wealth of tumor histology data, including anti-pimonidazole staining, no current methods use these data to induce a quantitative characterization of chronic tumor hypoxia in time and space. We use image-processing algorithms to develop a set of candidate image features that can formulate just such a quantitative description of xenographed colorectal chronic tumor hypoxia. Two features in particular give low-variance measures of chronic hypoxia near a vessel: intensity sampling that extends radially away from approximated blood vessel centroids, and multithresholding to segment tumor tissue into normal, hypoxic, and necrotic regions. From these features we derive a spatiotemporal logical expression whose truth value depends on its predicate clauses that are grounded in this histological evidence. As an alternative to the spatiotemporal logical formulation, we also propose a way to formulate a linear regression function that uses all of the image features to learn what chronic hypoxia looks like, and then gives a quantitative similarity score once it is trained on a set of histology images.

FindFoci: a focus detection algorithm with automated parameter training that closely matches human assignments, reduces human inconsistencies and increases speed of analysis

Accurate and reproducible quantification of the accumulation of proteins into foci in cells is essential for data interpretation and for biological inferences. To improve reproducibility, much emphasis has been placed on the preparation of samples, but less attention has been given to reporting and standardizing the quantification of foci. The current standard to quantitate foci in open-source software is to manually determine a range of parameters based on the outcome of one or a few representative images and then apply the parameter combination to the analysis of a larger dataset. Here, we demonstrate the power and utility of using machine learning to train a new algorithm (FindFoci) to determine optimal parameters. FindFoci closely matches human assignments and allows rapid automated exploration of parameter space. Thus, individuals can train the algorithm to mirror their own assignments and then automate focus counting using the same parameters across a large number of images. Using the training algorithm to match human assignments of foci, we demonstrate that applying an optimal parameter combination from a single image is not broadly applicable to analysis of other images scored by the same experimenter or by other experimenters. Our analysis thus reveals wide variation in human assignment of foci and their quantification. To overcome this, we developed training on multiple images, which reduces the inconsistency of using a single or a few images to set parameters for focus detection. FindFoci is provided as an open-source plugin for ImageJ.

Atomic force microscopic detection enabling multiplexed low-cycle-number quantitative polymerase chain reaction for biomarker assays

Quantitative polymerase chain reaction is the current "golden standard" for quantification of nucleic acids; however, its utility is constrained by an inability to easily and reliably detect multiple targets in a single reaction. We have successfully overcome this problem with a novel combination of two widely used approaches: target-specific multiplex amplification with 15 cycles of polymerase chain reaction (PCR), followed by single-molecule detection of amplicons with atomic force microscopy (AFM). In test experiments comparing the relative expression of ten transcripts in two different human total RNA samples, we find good agreement between our single reaction, multiplexed PCR/AFM data, and data from 20 individual singleplex quantitative PCR reactions. This technique can be applied to virtually any analytical problem requiring sensitive measurement concentrations of multiple nucleic acid targets.