HFM-Tracker: a cell tracking algorithm based on hybrid feature matching.

Cell migration is known to be a fundamental biological process, playing an essential role in development, homeostasis, and diseases. This paper introduces a cell tracking algorithm named HFM-Tracker (Hybrid Feature Matching Tracker) that automatically identifies cell migration behaviours in consecutive images. It combines Contour Attention (CA) and Adaptive Confusion Matrix (ACM) modules to accurately capture cell contours in each image and track the dynamic behaviors of migrating cells in the field of view. Cells are firstly located and identified via the CA module-based cell detection network, and then associated and tracked via a cell tracking algorithm employing a hybrid feature-matching strategy. This proposed HFM-Tracker exhibits superiorities in cell detection and tracking, achieving 75% in MOTA (Multiple Object Tracking Accuracy) and 65% in IDF1 (ID F1 score). It provides quantitative analysis of the cell morphology and migration features, which could further help in understanding the complicated and diverse cell migration processes.

SmartImage: A Machine Learning Method for Nanopore Identifying Chemical Modifications on RNA.

RNA modifications modulate essential cellular functions. However, it is challenging to quantitatively identify the differences in RNA modifications. To further improve the single-molecule sensing ability of nanopores, we propose a machine-learning algorithm called SmartImage for identifying and classifying nanopore electrochemical signals based on a combination of improved graph conversion methods and deep neural networks. SmartImage is effective for nearly all ranges of signal duration, which breaks the limitation of the current nanopore algorithm. The overall accuracy (OA) of our proposed recognition strategy exceeded 90% for identifying three types of RNAs. Prediction experiments show that the SmartImage owns the ability to recognize one modified RNA molecule from 1000 normal RNAs with OA >90%. Thus our proposed model and algorithm hold the potential application in clinical applications.

An ultra-low noise amplifier array system for high throughput single entity analysis.

Electrochemical measurements at the single entity level provide ultra-sensitive tools for the precise diagnosis and understanding of basic biological and chemical processes. By decoding current signatures, single-entity electrochemistry provides abundant information on charges, sizes, shapes, catalytic performances and compositions. The accuracy of single-entity electrochemistry highly relies on advanced instrumentation to achieve the amperometric resolution at the sub-picoampere level and the temporal resolution at the sub-microsecond level. Currently, it is still a challenge for paralleling amplifiers to allow low-noise and high bandwidth single-entity electrochemical measurements. Herein, we developed a low-noise four-channel electrochemical instrumentation that integrates an Au electrode array with amplifiers in the circuit board. With this amplifier array, we achieved a high bandwidth (>100 kHz) electrochemical measurement. The further practical experiments proved the capability of this amplifier array system in acquiring transient signals from both single-molecule detection with an aerolysin nanopore and single Pt nanoparticle catalysis during the dynamic collision process. Paired with appropriate microfluidic array systems, our instrumentation will enable an extraordinarily high-throughput feature for single-entity sensing.

Single-Molecule Frequency Fingerprint for Ion Interaction Networks in a Confined Nanopore.

The transport of molecules and ions through biological nanopores is governed by interaction networks among restricted ions, transported molecules, and residue moieties at pore inner walls. However, identification of such weak ion fluctuations from only few tens of ions inside nanopore is hard to achieve owing to electrochemical measurement limitations. Here, we developed an advanced frequency method to achieve qualitative and spectral analysis of ion interaction networks inside a nanopore. The peak frequency fm reveals the dissociation rate between nanopore and ions; the peak amplitude am depicts the amount of combined ions with the nanopore after interaction equilibrium. A mathematical model for single-molecule frequency fingerprint achieved the prediction of interaction characteristics of mutant nanopores. This single-molecule frequency fingerprint is important for classification, characterization, and prediction of synergetic interaction networks inside nanoconfinement.

Snapshotting the transient conformations and tracing the multiple pathways of single peptide folding using a solid-state nanopore.

A fundamental question relating to protein folding/unfolding is the time evolution of the folding of a protein into its precisely defined native structure. The proper identification of transition conformations is essential for accurately describing the dynamic protein folding/unfolding pathways. Owing to the rapid transitions and sub-nm conformation differences involved, the acquisition of the transient conformations and dynamics of proteins is difficult due to limited instrumental resolution. Using the electrochemical confinement effect of a solid-state nanopore, we were able to snapshot the transient conformations and trace the multiple transition pathways of a single peptide inside a nanopore. By combining the results with a Markov chain model, this new single-molecule technique is applied to clarify the transition pathways of the ß-hairpin peptide, which shows nonequilibrium fluctuations among several blockage current stages. This method enables the high-throughput investigation of transition pathways experimentally to access previously obscure peptide dynamics, which is significant for understanding the folding/unfolding mechanisms and misfolding of peptides or proteins.

Segmentation of retinal fluid based on deep learning: application of three-dimensional fully convolutional neural networks in optical coherence tomography images.

AIM: To explore a segmentation algorithm based on deep learning to achieve accurate diagnosis and treatment of patients with retinal fluid. METHODS: A two-dimensional (2D) fully convolutional network for retinal segmentation was employed. In order to solve the category imbalance in retinal optical coherence tomography (OCT) images, the network parameters and loss function based on the 2D fully convolutional network were modified. For this network, the correlations of corresponding positions among adjacent images in space are ignored. Thus, we proposed a three-dimensional (3D) fully convolutional network for segmentation in the retinal OCT images. RESULTS: The algorithm was evaluated according to segmentation accuracy, Kappa coefficient, and F1 score. For the 3D fully convolutional network proposed in this paper, the overall segmentation accuracy rate is 99.56%, Kappa coefficient is 98.47%, and F1 score of retinal fluid is 95.50%. CONCLUSION: The OCT image segmentation algorithm based on deep learning is primarily founded on the 2D convolutional network. The 3D network architecture proposed in this paper reduces the influence of category imbalance, realizes end-to-end segmentation of volume images, and achieves optimal segmentation results. The segmentation maps are practically the same as the manual annotations of doctors, and can provide doctors with more accurate diagnostic data.

Measuring a frequency spectrum for single-molecule interactions with a confined nanopore.

Nanopore analysis is a powerful technique for single molecule analysis by virtue of its electrochemically confined effects. As a single molecule translocates through the nanopore, the featured ionic current pattern on the time scale contains single molecule characteristics including volume, charge, and conformational properties. Although the characteristics of a single molecule in a nanopore have been written to the featured ionic current, extracting the dynamic information from a complex current trace is still a big challenge. Here, we present an applicable nanopore analysis method employing the Hilbert-Huang Transform (HHT) to study the vibrational features and interactions of a single molecule during the dynamic translocation process through the confined space of a nanopore. The HHT method is specially developed for analyzing nonlinear and non-stationary data that is highly compatible with nanopore data with a high frequency resolution. To provide proof-of-concept, we applied HHT to measure the frequency response for the wild-type (WT) aerolysin and mutant K238E aerolysin nanopores with and without the presence of poly(dA)4, respectively. The energy-frequency-time distribution spectra demonstrate that the biological nanopore contributes greatly to the characteristics of the high frequency component (>2 kHz) in the current recording. Our results suggest that poly(dA)4 undergoes relatively more consistent and confined interactions with K238E than WT, leading to a prolonging of the duration time. Therefore, the characteristics in frequency analysis could be regarded as an "single-molecule ionic spectrum" inside the nanopore, which encodes the detailed behaviours of single-molecule weak interactions.