Deep learning-based classification of breast cancer cells using transmembrane receptor dynamics.

MOTIVATION: Motions of transmembrane receptors on cancer cell surfaces can reveal biophysical features of the cancer cells, thus providing a method for characterizing cancer cell phenotypes. While conventional analysis of receptor motions in the cell membrane mostly relies on the mean-squared displacement plots, much information is lost when producing these plots from the trajectories. Here we employ deep learning to classify breast cancer cell types based on the trajectories of epidermal growth factor receptor (EGFR). Our model is an artificial neural network trained on the EGFR motions acquired from six breast cancer cell lines of varying invasiveness and receptor status: MCF7 (hormone receptor positive), BT474 (HER2-positive), SKBR3 (HER2-positive), MDA-MB-468 (triple negative, TN), MDA-MB-231 (TN) and BT549 (TN). RESULTS: The model successfully classified the trajectories within individual cell lines with 83% accuracy and predicted receptor status with 85% accuracy. To further validate the method, epithelial-mesenchymal transition (EMT) was induced in benign MCF10A cells, noninvasive MCF7 cancer cells and highly invasive MDA-MB-231 cancer cells, and EGFR trajectories from these cells were tested. As expected, after EMT induction, both MCF10A and MCF7 cells showed higher rates of classification as TN cells, but not the MDA-MB-231 cells. Whereas deep learning-based cancer cell classifications are primarily based on the optical transmission images of cell morphology and the fluorescence images of cell organelles or cytoskeletal structures, here we demonstrated an alternative way to classify cancer cells using a dynamic, biophysical feature that is readily accessible. AVAILABILITY AND IMPLEMENTATION: A python implementation of deep learning-based classification can be found at https://github.com/soonwoohong/Deep-learning-for-EGFR-trajectory-classification. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Measuring DNA Hybridization Kinetics in Live Cells Using a Time-Resolved 3D Single-Molecule Tracking Method.

Single-molecule detection enables direct characterization of annealing/melting kinetics of nucleic acids without the need for synchronization of molecular states, but the current experiments are not carried out in a native cellular context. Here we describe an integrated 3D single-molecule tracking and lifetime measurement method that can follow individual DNA molecules diffusing inside a mammalian cell and observe multiple annealing and melting events on the same molecules. By comparing the hybridization kinetics of the same DNA strand in vitro, we found the association constants can be 13- to 163-fold higher in the molecular crowding cellular environment.

LNA Thymidine Monomer Enables Differentiation of the Four Single-Nucleotide Variants by Melting Temperature.

High-resolution melting (HRM) analysis of DNA is a closed-tube single-nucleotide polymorphism (SNP) detection method that has shown many advantages in point-of-care diagnostics and personalized medicine. While recently developed melting probes have demonstrated significantly improved discrimination of mismatched (mutant) alleles from matched (wild-type) alleles, no effort has been made to design a simple melting probe that can reliably distinguish all four SNP alleles in a single experiment. Such a new probe could facilitate the discovery of rare genetic mutations at lower cost. Here we demonstrate that a melting probe embedded with a single locked thymidine monomer (tL) can reliably differentiate the four SNP alleles by four distinct melting temperatures (termed the "4Tm probe"). This enhanced discriminatory power comes from the decreased melting temperature of the tL·C mismatched hybrid as compared to that of the t·C mismatched hybrid, while the melting temperatures of the tL-A, tL·G and tL·T hybrids are increased or remain unchanged as compared to those of their canonical counterparts. This phenomenon is observed not only in the HRM experiments but also in the molecular dynamics simulations.

Segmentation of 3D Trajectories Acquired by TSUNAMI Microscope: An Application to EGFR Trafficking.

Whereas important discoveries made by single-particle tracking have changed our view of the plasma membrane organization and motor protein dynamics in the past three decades, experimental studies of intracellular processes using single-particle tracking are rather scarce because of the lack of three-dimensional (3D) tracking capacity. In this study we use a newly developed 3D single-particle tracking method termed TSUNAMI (Tracking of Single particles Using Nonlinear And Multiplexed Illumination) to investigate epidermal growth factor receptor (EGFR) trafficking dynamics in live cells at 16/43 nm (xy/z) spatial resolution, with track duration ranging from 2 to 10 min and vertical tracking depth up to tens of microns. To analyze the long 3D trajectories generated by the TSUNAMI microscope, we developed a trajectory analysis algorithm, which reaches 81% segment classification accuracy in control experiments (termed simulated movement experiments). When analyzing 95 EGF-stimulated EGFR trajectories acquired in live skin cancer cells, we find that these trajectories can be separated into three groups-immobilization (24.2%), membrane diffusion only (51.6%), and transport from membrane to cytoplasm (24.2%). When EGFRs are membrane-bound, they show an interchange of Brownian diffusion and confined diffusion. When EGFRs are internalized, transitions from confined diffusion to directed diffusion and from directed diffusion back to confined diffusion are clearly seen. This observation agrees well with the model of clathrin-mediated endocytosis.

Single-Molecule Tracking and Its Application in Biomolecular Binding Detection.

In the past two decades significant advances have been made in single-molecule detection, which enables the direct observation of single biomolecules at work in real time and under physiological conditions. In particular, the development of single-molecule tracking (SMT) microscopy allows us to monitor the motion paths of individual biomolecules in living systems, unveiling the localization dynamics and transport modalities of the biomolecules that support the development of life. Beyond the capabilities of traditional camera-based tracking techniques, state-of-the-art SMT microscopies developed in recent years can record fluorescence lifetime while tracking a single molecule in the 3D space. This multiparameter detection capability can open the door to a wide range of investigations at the cellular or tissue level, including identification of molecular interaction hotspots and characterization of association/dissociation kinetics between molecules. In this review, we discuss various SMT techniques developed to date, with an emphasis on our recent development of the next generation 3D tracking system that not only achieves ultrahigh spatiotemporal resolution but also provides sufficient working depth suitable for live animal imaging. We also discuss the challenges that current SMT techniques are facing and the potential strategies to tackle those challenges.

NanoCluster Beacons Enable Detection of a Single N6-Methyladenine.

While N(6)-methyladenine (m(6)A) is a common modification in prokaryotic and lower eukaryotic genomes and has many biological functions, there is no simple and cost-effective way to identify a single N(6)-methyladenine in a nucleic acid target. Here we introduce a robust, simple, enzyme-free and hybridization-based method using a new silver cluster probe, termed methyladenine-specific NanoCluster Beacon (maNCB), which can detect single m(6)A in DNA targets based on the fluorescence emission spectra of silver clusters. Not only can maNCB identify m(6)A at the single-base level but it also can quantify the extent of adenine methylation in heterogeneous samples. Our method is superior to high-resolution melting analysis as we can pinpoint the location of m(6)A in the target.

Utilization of multi-band OFDM modulation to increase traffic rate of phosphor-LED wireless VLC.

To increase the traffic rate in phosphor-LED visible light communication (VLC), a multi-band orthogonal frequency division multiplexed (OFDM) modulation is first proposed and demonstrated. In the measurement, we do not utilize optical blue filter to increase modulation bandwidth of phosphor-LED in the VLC system. In this proposed scheme, different bands of OFDM signals are applied to different LED chips in a LED lamp, this can avoid the power fading and nonlinearity issue by applying the same OFDM signal to all the LED chips in a LED lamp. Here, the maximum increase percentages of traffic rates are 41.1%, 17.8% and 17.8% under received illuminations of 200, 500 and 1000 Lux, respectively, when the proposed three-band OFDM modulation is used in the VLC system. In addition, the analysis and verification by experiments are also performed.

What Is the Rerevision Rate After Revising a Hip Resurfacing Arthroplasty? Analysis From the AOANJRR.

BACKGROUND: More than 15,000 primary hip resurfacing arthroplasties have been recorded by the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) with 884 primary procedures requiring revision for reasons other than infection, a cumulative percent revision rate at 12 years of 11%. However, few studies have reported the survivorship of these revision procedures. QUESTIONS/PURPOSES: (1) What is the cumulative percent rerevision rate for revision procedures for failed hip resurfacings? (2) Is there a difference in rerevision rate among different types of revision or bearing surfaces? METHODS: The AOANJRR collects data on all primary and revision hip joint arthroplasties performed in Australia and after verification against health department data, checking of unmatched procedures, and subsequent retrieval of unreported procedures is able to obtain an almost complete data set relating to hip arthroplasty in Australia. Revision procedures are linked to the known primary hip arthroplasty. There were 15,360 primary resurfacing hip arthroplasties recorded of which 884 had undergone revision and this was the cohort available to study. The types of revisions were acetabular only, femoral only, or revision of both acetabular and femoral components. With the exception of the acetabular-only revisions, all revisions converted hip resurfacing arthroplasties to conventional (stemmed) total hip arthroplasties (THAs). All initial revisions for infection were excluded. The survivorship of the different types of revisions and that of the different bearing surfaces used were estimated using the Kaplan-Meier method and compared using Cox proportional hazard models. Cumulative percent revision was calculated by determining the complement of the Kaplan-Meier survivorship function at that time multiplied by 100. RESULTS: Of the 884 revisions recorded, 102 underwent further revision, a cumulative percent rerevision at 10 years of 26% (95% confidence interval, 19.6-33.5). There was no difference in the rate of rerevision between acetabular revision and combined femoral and acetabular revision (hazard ratio [HR], 1.06 [0.47-2], p = 0.888), femoral revision and combined femoral and acetabular revision (HR, 1.00 [0.65-2], p = 0.987), and acetabular revision and femoral revision (HR, 1.06 [0.47-2], p = 0.893). There was no difference in the rate of rerevision when comparing different bearing surfaces (metal-on-metal versus ceramic-on-ceramic HR, 0.46 [0.16-1.29], p = 0.141; metal-on-metal versus ceramic-on-crosslinked polyethylene HR, 0.51 [0.15-1.76], p = 0.285; metal-on-metal versus metal-on-crosslinked polyethylene HR, 0.62 [0.20-1.89], p = 0.399; and metal-on-metal versus oxinium-on-crosslinked polyethylene HR, 0.53 [0.14-2.05], p = 0.356). CONCLUSIONS: Revision of a primary hip resurfacing arthroplasty is associated with a high risk of rerevision. This study may help surgeons guide their patients about the outcomes in the longer term after the first revision of hip resurfacing arthroplasty. LEVEL OF EVIDENCE: Level III, therapeutic study.

Higher Rate of Revision in PFC Sigma Primary Total Knee Arthroplasty With Mismatch of Femoro-Tibial Component Sizes.

Total knee arthroplasty (TKA) systems permit a degree of femoro-tibial component size mismatch. The effect of mismatched components on revision rates has not been evaluated in a large study. We reviewed 21,906 fixed-bearing PFC Sigma primary TKAs using the Australian Orthopaedic Association National Joint Replacement Registry, dividing patients into three groups: no femoro-tibial size mismatch, tibial component size > femoral component size, and femoral component > tibial component. Revision rates were higher when the femoral size was greater than the tibia, compared to both equal size (HR = 1.20 (1.00, 1.45), P = 0.047) and to tibial size greater than femoral (HR = 1.60 (1.08, 2.37), P = 0.019). Potential mechanisms to explain these findings include edge loading of polyethylene and increased tibial component stresses.

The utilization of incinerated hip and knee prostheses for identification.

PURPOSE: The aim of this study was to test various methods of retrieving number data from hip and knee implants from cremated human remains and to validate our findings by cross referencing our results with the national joint replacement registry. METHOD: Implants were collected from the remains of individuals who had donated their bodies to science following routine planned cremation. A number of different chemical and physical methods to expose the implant numbers on cremated implants to the point that they were legible were tested. The retrieved data on the implants was referred to the Australian Orthopaedic Association National Joint Replacement Registry to identify the individuals, and the names were cross-referenced from the original list of donors. RESULTS: It was possible to retrieve sufficient data from cremated implants to track the name of the recipient of implants if they were placed following the formation of the registry. Both wet and dry paper (1200 size and without moisture), and fine grade steel wool (used in antique restoration), were successful in removing the oxidized layer from implants. With hip implants, it was discovered that the best area to retrieve clear readable information is inside the ball head or at the end of the neck as this area is protected from oxidation during incineration. CONCLUSION: Incinerated or cremated hip and knee implants may be used to assist in the identification of a decedent following careful treatment, in conjunction with national joint revision registries and company data.

A biomimetic honeycomb-like scaffold prepared by flow-focusing technology for cartilage regeneration.

A tissue engineering chondrocytes/scaffold construct provides a promise to cartilage regeneration. The architecture of a scaffold such as interconnections, porosities, and pore sizes influences the fates of seeding cells including gene expression, survival, migration, proliferation, and differentiation thus may determine the success of this approach. Scaffolds of highly ordered and uniform structures are desirable to control cellular behaviors. In this study, a newly designed microfluidic device based on flow-focusing geometry was developed to fabricate gelatin scaffolds of ordered pores. In comparison with random foam scaffolds made by the conventional freeze-dried method, honeycomb-like scaffolds exhibit higher swelling ratio, porosity, and comparable compressive strength. In addition, chondrocytes grown in the honeycomb-like scaffolds had good cell viability, survival rate, glycosaminoglycans production, and a better proliferation than ones in freeze-dried scaffolds. Real-time PCR analysis showed that the mRNA expressions of aggrecan and collagen type II were up-regulated when chondrocytes cultured in honeycomb-like scaffolds rather than cells cultured as monolayer fashion. Oppositely, chondrocytes expressed collagen type II as monolayer culture when seeded in freeze-dried scaffolds. Histologic examinations revealed that cells produced proteoglycan and distributed uniformly in honeycomb-like scaffolds. Immunostaining showed protein expression of S-100 and collagen type II but negative for collagen type I and X, which represents the chondrocytes maintained normal phenotype. In conclusion, a highly ordered and honeycomb-like scaffold shows superior performance in cartilage tissue engineering. Biotechnol. Bioeng. 2014;111: 2338-2348. © 2014 Wiley Periodicals, Inc.

Real-time white-light phosphor-LED visible light communication (VLC) with compact size.

In this demonstration, we first demonstrate a real-time phosphor-LED visible light communication (VLC) system with 37 Mbit/s total throughput under a 1.5 m free space transmission length. The transmitter and receiver modules are compact size. Utilizing our proposed pre-equalization technology, the ~1 MHz bandwidth of phosphor LED could be extended to ~12 MHz without using blue filter. Thus, the increase in bandwidth would enhance the traffic data rate for VLC transmission. The maximum bit-rate achieved by the VLC system is 37 Mbit/s, and a video transmission at 28.419 Mbit/s is demonstrated using the proposed VLC system. In addition, the relationships of received power and signal performance are discussed and analyzed.

Optical zoom module based on two deformable mirrors for mobile device applications.

In recent years, optical zoom functionality in mobile devices has been studied. Traditional zoom systems use motors to change separation of lenses to achieve the zoom function, but these systems result in long total length and high power consumption, which are not suitable for mobile devices. Adopting micromachined polymer deformable mirrors in zoom systems has the potential to reduce thickness and chromatic aberration. In this paper, we propose a 2× continuous optical zoom system with five-megapixel image sensors by using two deformable mirrors. In our design, the thickness of the zoom system is about 11 mm. The effective focal length ranges from 4.7 mm at a field angle of 52° to 9.4 mm. The f-number is 4.4 and 6.4 at the wide-angle and telephoto end, respectively.

Generative adversarial network enables rapid and robust fluorescence lifetime image analysis in live cells.

Fluorescence lifetime imaging microscopy (FLIM) is a powerful tool to quantify molecular compositions and study molecular states in complex cellular environment as the lifetime readings are not biased by fluorophore concentration or excitation power. However, the current methods to generate FLIM images are either computationally intensive or unreliable when the number of photons acquired at each pixel is low. Here we introduce a new deep learning-based method termed flimGANE (fluorescence lifetime imaging based on Generative Adversarial Network Estimation) that can rapidly generate accurate and high-quality FLIM images even in the photon-starved conditions. We demonstrated our model is up to 2,800 times faster than the gold standard time-domain maximum likelihood estimation (TD_MLE) and that flimGANE provides a more accurate analysis of low-photon-count histograms in barcode identification, cellular structure visualization, Förster resonance energy transfer characterization, and metabolic state analysis in live cells. With its advantages in speed and reliability, flimGANE is particularly useful in fundamental biological research and clinical applications, where high-speed analysis is critical.

Atelocollagen-Embedded Chondrocyte Precursors as a Treatment for Grade-4 Cartilage Defects of the Femoral Condyle: A Case Series with up to 9-Year Follow-Up.

We demonstrated the safety and efficacy of autologous chondrocyte precursor (CP) cell therapy in repairing Grade 4 cartilage defects of medial femoral condyles. The autologous bone marrow mesenchymal stem cells of each participant were isolated, amplified, and then differentiated into CPs in atelocollagen. Neotissues made of CPs were implanted into cartilage defects with an average cell density of 4.9 ± 2.1 × 106 cells/cm2 through arthrotomy. The knee function was evaluated with the International Knee Documentation Committee (IKDC) subjective knee form. Patients' knee functions significantly improved by the 28th week (IKDC score = 68.3 ± 12.1), relative to the initial functionality before the CP therapy (IKDC score = 46.1 ± 16.4, p-value = 0.0014). Nine of these twelve patients maintained good knee functions for 9 years post-implantation (IKDC score = 69.8 ± 12.3) at levels higher than the pre-implantation values (p-value = 0.0018). Patients were evaluated with MRI and arthroscopy, and the defective sites exhibited a smooth surface without a gap between the implant and host tissue. This study demonstrates that autologous CPs successfully engraft into the host tissue and result in the re-formation of hyaline-like cartilage, thereby improving the impaired knee functions. Most importantly, no adverse event was reported during this long-term follow-up period.

Safety and Efficacy of Kartigen® in Treating Cartilage Defects: A Randomized, Controlled, Phase I Trial.

Here, we aimed to investigate the safety and preliminary efficacy of Kartigen®, a matrix with autologous bone marrow mesenchymal stem cell-derived chondrocyte precursors embedded in atelocollagen. As a surgical graft, Kartigen® was implanted onto the cartilage defects at the weight-bearing site of the medial femoral condyle of the knee. Fifteen patients were enrolled and stratified into two groups, undergoing either Kartigen® implantation (n = 10) or microfracture (control group, n = 5). The primary endpoint was to evaluate the safety of Kartigen® by monitoring the occurrence of adverse events through physician queries, physical examinations, laboratory tests, and radiological analyses for 2 years. There were no infections, inflammations, adhesions, loose body, or tumor formations in the Kartigen®-implanted knees. The preliminary efficacy was assessed using the International Knee Documentation Committee (IKDC) score, visual analog scale, and second-look arthroscopy. The postoperative IKDC scores of the Kartigen® group significantly improved in the 16th week (IKDC = 62.1 ± 12.8, p = 0.025), kept increasing in the first year (IKDC = 78.2 ± 15.4, p < 0.005), and remained satisfactory in the second year (IKDC = 73.6 ± 13.8, p < 0.005), compared to the preoperative condition (IKDC = 47.1 ± 17.0), while the postoperative IKDC scores of the control group also achieved significant improvement in the 28th week (IKDC = 68.5 ± 6.1, p = 0.032) versus preoperative state (IKDC = 54.0 ± 9.1). However, the IKDC scores decreased in the first year (IKDC = 63.5 ± 11.6) as well as in the second year (IKDC = 52.6 ± 16.4). Thirteen patients underwent second-look arthroscopy and biopsy one year after the operation. The Kartigen® group exhibited integration between Kartigen® and host tissue with a smooth appearance at the recipient site, whereas the microfracture group showed fibrillated surfaces. The histological and immunohistochemical analyses of biopsy specimens demonstrated the columnar structure of articular cartilage and existence of collagen type II and glycosaminoglycan mimic hyaline cartilage. This study indicates that Kartigen® is safe and effective in treating cartilage defects.

Three-Dimensional Two-Color Dual-Particle Tracking Microscope for Monitoring DNA Conformational Changes and Nanoparticle Landings on Live Cells.

Here, we present a three-dimensional two-color dual-particle tracking (3D-2C-DPT) technique that can simultaneously localize two spectrally distinct targets in three dimensions with a time resolution down to 5 ms. The dual-targets can be tracked with separation distances from 33 to 250 nm with tracking precisions of ∼15 nm (for static targets) and ∼35 nm (for freely diffusing targets). Since each target is individually localized, a wealth of data can be extracted, such as the relative 3D position, the 2D rotation, and the separation distance between the two targets. Using this technique, we turn a double-stranded DNA (dsDNA)-linked dumbbell-like dimer into a nanoscopic optical ruler to quantify the bending dynamics of nicked or gapped dsDNA molecules in free solution by manipulating the design of dsDNA linkers (1-nick, 3-nt, 6-nt, or 9-nt single-strand gap), and the results show the increase of kon (linear to bent) from 3.2 to 10.7 s-1. The 3D-2C-DPT is then applied to observe translational and rotational motions of the landing of an antibody-conjugated nanoparticle on the plasma membrane of living cells, revealing the reduction of rotations possibly due to interactions with membrane receptors. This study demonstrates that this 3D-2C-DPT technique is a new tool to shed light on the conformational changes of biomolecules and the intermolecular interactions on plasma membrane.

Digital Receptor Occupancy Assay in Quantifying On- and Off-Target Binding Affinities of Therapeutic Antibodies.

While monoclonal antibodies are the fastest-growing class of therapeutic agents, we lack a method that can directly quantify the on- and off-target binding affinities of newly developed therapeutic antibodies in crude cell lysates. As a result, some therapeutic antibody candidates could have a moderate on-target binding affinity but a high off-target binding affinity, which not only gives a reduced efficacy but triggers unwanted side effects. Here, we report a single-molecule counting method that precisely quantifies antibody-bound receptors, free receptors, and unbound antibodies in crude cell lysates, termed digital receptor occupancy assay (DRO). Compared to the traditional flow cytometry-based binding assay, DRO assay enables direct and digital quantification of the three molecular species in solution without the additional antibodies for competitive binding. When characterizing the therapeutic antibody, cetuximab, using DRO assay, we found the on-target binding ratio to be 65% and the binding constant (Kd) to be 2.4 nM, while the off-target binding causes the binding constant to decrease by 0.3 nM. Other than cultured cells, the DRO assay can be performed on tumor mouse xenograft models. Thus, DRO is a simple and highly quantitative method for cell-based antibody binding analysis which can be broadly applied to screen and validate new therapeutic antibodies.

Polymer dots enable deep in vivo multiphoton fluorescence imaging of microvasculature.

Deep in vivo imaging of vasculature requires small, bright, and photostable fluorophores suitable for multiphoton microscopy (MPM). Although semiconducting polymer dots (pdots) are an emerging class of highly fluorescent contrast agents with favorable advantages for the next generation of in vivo imaging, their use for deep MPM has never before been demonstrated. Herein, we characterize the multiphoton properties of three pdot variants and perform deep in vivo MPM imaging of cortical rodent microvasculature. We find pdot brightness exceeds conventional fluorophores, including quantum dots, and their broad multiphoton absorption spectrum permits imaging at wavelengths better-suited for biological imaging and confers compatibility with a range of longer excitation wavelengths. This results in substantial improvements in signal-to-background ratio (>3.5-fold) and greater cortical imaging depths (z = 1,300 µm). Ultimately, pdots are a versatile tool for MPM due to their extraordinary brightness and broad absorption, enabling interrogation of deep structures in vivo.

Improving NanoCluster Beacon performance by blocking the unlabeled NC probes.

While NanoCluster Beacon (NCB) is a versatile molecular probe, it suffers from a low target-specific signal issue due to impurities. Here we show that adding a "blocker" strand to the reaction can effectively block the nonfunctional probes and enhance the target-specific signal by 14 fold at a 0.1 target/probe ratio.