Biology subjects

Rollins, D. E.

Publications and source records attributed to Rollins, D. E..

4 recordsLinked to original sources

bioRxiv2026

afMLevel: Deep learning U-Net Models for Levelling Atomic Force Microscopy Images and Movies

AFM experiments can now produce hundreds of frames per session and this rises to thousands in the case of video-rate high-speed AFM, but every image must still be levelled to remove tilt, bowing and line noise before it can be measured. Current automatic routines rely on user-chosen masks and parameters that work for some sample types but fail on others, so levelling often remains a manual, subjective bottleneck. Here, we present two U-Net models developed using machine learning for levelling AFM images without altering the sample topography. MLMask segments features so that they are excluded from conventional plane and line background fits, and MLBackground predicts the noise and artifact background directly, which is then subtracted. Both were trained on 2,001 experimental images from multiple instruments and sample types, using manually levelled images as ground truth. On 164 unseen images, MLMask routines brought 70-80% of images below a mean squared error of 0.1 relative to manual levelling, compared with 60-70% for the best multi-plane method, with a lower mean error (0.1-0.15 vs 0.15-0.3), and both models significantly outperformed fourteen classical routines (p < 0.01). Applied frame by frame, the models process 512x512 pixel HS-AFM movies at 4.7 frames.s-1. We provide afMLevel as an openly-available package for the use of other researchers. By removing the need to choose levelling parameters image by image, this package together with its models and paired training data, provides a foundation for automated AFM analysis pipelines.

biophysics↗

bioRxiv2025

Imaging and mechanism of DNA-DNA recognition mediated by divalent ions

In the cell, DNA must be tightly packed to facilitate its organisation into the nucleus, where recognition of homologous sequences underpins key processes such as recombination. Yet the structural basis of DNA-DNA pairing remains unknown. Here we combine high-resolution AFM and atomistic MD simulations to provide the first direct visualisation of DNA pairing in the presence of divalent ions. We show that strongly paired DNAs often achieve groove-to-groove alignment, driven by ionic bridges connecting the minor grooves of the two duplexes. These contacts are further stabilized by sequence-specific interactions, whose strength and specificity vary with the ion type. This mechanism of ion-mediated groove alignment provides a molecular framework for the long-hypothesized "helical alignment" model, in which homologous recognition is facilitated by preserving structural register between the two helices. Together, our findings reveal a fundamental principle by which divalent ions promote DNA-DNA pairing, with broad implications for chromosomal condensation and genome organization.

biophysics↗

bioRxiv2024

Multiscale topological analysis of kinetoplast DNA via high-resolution AFM

Kinetoplast DNA is a complex nanoscale network, naturally assembled from thousands of interconnected DNA circles within the mitochondrion of certain parasites. Despite the relevance of this molecule to parasitology and the recent discovery of tuneable mechanics, its topology remains highly contested. Here we present a multiscale analysis into the structure of kDNA using a combination of high-resolution atomic force microscopy and custom-designed image analysis protocols. By capturing a notably large set of high-resolution images, we are able to look beyond individual kDNA variations and quantify population properties throughout several length scales. Within the sample, geometric fluctuations of area and mean curvature are observed, corresponding with previous in-vitro measurements. These translate to localised variations in density, with a sample-wide decrease in DNA density from the outer rim of the molecule to the centre and an increase in pore size. Nodes were investigated in a single molecule study, and their estimated connectivity significantly exceeded mean valence, with a high dependence on their position in the network. While node separation was approximately half the minicircle circumference, it followed a strong bimodal distribution, suggesting more complex underlying behaviour. Finally, upon selective digestion of the network, breakdown of the fibril-cap heterogeneity was observed, with molecules expanding less upon immobilisation on the mica surface. Additionally, selective digestion was seen in localised areas of the network, increasing pore size disproportionately. Overall, the combination of high-resolution AFM and single molecule image analysis provides a promising method to the continued investigation of complex nanoscale structures. These findings support the ongoing characterisation of kDNA topology to aid understanding of its biological and mechanical phenomena.

biophysics↗

bioRxiv2022

Autophagy receptor NDP52 alters DNA conformation to modulate RNA Polymerase II transcription

NDP52 is an autophagy receptor involved in the recognition and degradation of invading pathogens and damaged organelles. Although NDP52 was first identified in the nucleus and is expressed throughout the cell, to date, there is no clear nuclear function for NDP52. Here, we use a multidisciplinary approach to characterise the biochemical properties and nuclear roles of NDP52. We found that NDP52 clusters with RNA Polymerase II (RNAPII) at transcription initiation sites and that its overexpression promotes the formation of additional transcriptional clusters. We also show that depletion of NDP52 impacts overall gene-expression levels in two model mammalian cells, and that transcription inhibition affects the spatial organisation and molecular dynamics of NDP52 in the nucleus. This directly links NDP52 to a role in RNAPII-dependent transcription. Furthermore, we also show that NDP52 binds specifically and with high affinity to double-stranded DNA (dsDNA) and that this interaction leads to changes in DNA structure in vitro. This, together with our proteomics data indicating enrichment for interactions with nucleosome remodelling proteins and DNA structure regulators, suggests a possible function for NDP52 in chromatin regulation. Overall, here we uncover novel nuclear roles for NDP52 in gene expression and DNA structure regulation.

biophysics↗