Label2label: Training a neural network to selectively restore cellular structures in fluorescence microscopy

Immunofluorescence (IF) microscopy is routinely used to visualise the spatial distribution of proteins that dictates their cellular function. However, unspecific antibody binding often results in high cytosolic background signals, decreasing the image contrast of a target structure. Recently, convolutional neural networks (CNNs) were successfully employed for image restoration in IF microscopy, but current methods cannot correct for those background signals. We report a new method that trains a CNN to reduce unspecific signals in IF images; we name this method label2label (L2L). In L2L, a CNN is trained with image pairs of two non-identical labels that target the same cellular structure. We show that after L2L training a network predicts images with significantly increased contrast of a target structure, which is further improved after implementing a multi-scale structural similarity loss function. Here, our results suggest that sample differences in the training data decrease hallucination effects that are observed with other methods. We further assess the performance of a cycle generative adversarial network, and show that a CNN can be trained to separate structures in superposed IF images of two targets.

Locus Coeruleus Optogenetic Modulation: Lessons Learned from Temporal Patterns

After reviewing seminal studies using optogenetics to interrogate the functional role of the locus coeruleus in behavior, we conclude that differences in firing rates and firing patterns of locus coeruleus neurons contribute to locus coeruleus nucleus heterogeneity by recruiting different output circuitry, and differentially modifying behavior. The outcomes initiated by different optogenetic input activation patterns and frequencies can have opposite consequences for behavior, activate different neurons in the same target structure, be supported by distinct adrenoceptors and vary with behavioral state.

What drives learner errors in written L2 production?

Accuracy in written L2 production can be influenced by many factors, including: (a) the relative similarity of the target structure to equivalent structure in the learner’s L1, and (b) the complexity of the target structure itself. The question of which of these two factors plays a stronger role is fundamental to theories of L2 acquisition. This written learner corpus study uses the English genitive alternation – s-genitives (‘the country's future’) and of-genitives (‘the future of the country’) – to attempt to shed light on this issue. L1 Spanish speakers lag behind L1 Japanese speakers in terms of accuracy rates when the target structure is an s-genitive. This L1 influence appears secondary to structural complexity effects; learners in both groups consistently use the simpler of-genitive with far higher accuracy. Both L1 and complexity effects are stronger in plural possessor contexts, with the plural feature apparently exacerbating learner difficulties with the s-genitive.

Programming cell growth into different cluster shapes using diffusible signals

Advances in genetic engineering technologies have allowed the construction of artificial genetic circuits, which have been used to generate spatial patterns of differential gene expression. However, the question of how cells can be programmed, and how complex the rules need to be, to achieve a desired tissue morphology has received less attention. Here, we address these questions by developing a mathematical model to study how cells can collectively grow into clusters with different structural morphologies by secreting diffusible signals that can influence cellular growth rates. We formulate how growth regulators can be used to control the formation of cellular protrusions and how the range of achievable structures scales with the number of distinct signals. We show that a single growth inhibitor is insufficient for the formation of multiple protrusions but may be achieved with multiple growth inhibitors, and that other types of signals can regulate the shape of protrusion tips. These examples illustrate how our approach could potentially be used to guide the design of regulatory circuits for achieving a desired target structure.

Observation of Clinical Efficacy of Parylene-Coated Occluder Based on Electronic Image of Visual Sensor in the Treatment of Congenital Heart Disease

With the acceleration of people’s life rhythm, the incidence of congenital heart disease is getting higher and higher. This study mainly explores the clinical efficacy observation of parylene-coated occluders based on the electronic image of the visual sensor in the treatment of congenital heart disease. The patient takes a supine position and observes whether the target structure (left ventricular cavity) information is separated from the capture area during rotation, and the angle of the probe is adjusted to ensure that as much target structure information as possible can be obtained during the two-dimensional image acquisition. During the three-dimensional ultrasound sampling process, a total of 60 coaxial images were collected within the range of 180°. To reflect the effect of virtual reality, it is necessary to provide users with interactive operation tools. This research provides two methods, mouse interaction and keyboard interaction. Through these two methods, the effect of the basic virtual endoscope is realized. Xtion’s image registration process is completed inside its PS1080 system-level chip, and all sensor information, color images, and depth images are transmitted to the user host through the USB interface. For example, the point of view can be placed in the heart, because the simulation occlusion must first find a suitable section to better observe and judge the occlusion effect. Therefore, this study uses the mouse interaction method to change the cutting by dragging and rotating the mouse. In the process of releasing the occluder, the parylene-coated occluder can be released after the X-ray and ultrasound examination results are satisfactory, and the delivery long sheath and catheter can be sprinkled and compressed to stop the bleeding. During the operation, the ASD measured by two-dimensional ultrasound after balloon inflation was 24.17 ± 7.94 mm, and the anterior and posterior diameter was 21.30 ± 6.54 mm. This research is helpful for the qualitative and quantitative diagnosis of congenital heart disease.

ADT2FHIR – A Tool for Converting ADT/GEKID Oncology Data to HL7 FHIR Resources

Harmonized and interoperable data management is a core requirement for federated infrastructures in clinical research. Institutions participating in such infrastructures often have to invest large degrees of time and resources in implementing necessary data integration processes to convert their local data to the required target structure. If the data is already available in an alternative shared data structure, the transformation from source to the desired target structure can be implemented once and then be distributed to all participants to reduce effort and harmonize results. The HL7® FHIR® standard is used as a basis for the shared data model of several medical consortia like DKTK and GBA. It is based on so-called resources which can be represented in XML. Oncological data in German university hospitals is commonly available in the ADT/GEKID format. From this common basis we conceptualized and implemented a transformation which accepts ADT/GEKID XML files and returns FHIR resources. We identified several problems with using the general ADT/GEKID structure in federated research infrastructures, as well as some possible pitfalls relating to the FHIR need for resource ids and focus on semantic coding which differs from the approach in the ADT/GEKID standard. To facilitate participation in federated infrastructures, we propose the ADT2FHIR transformation tool for partners with oncological data in the ADT/GEKID format.

Graph convolutional network for protein sequence sampling and prediction using structural data

The increasing integration between protein engineering and machine learning has led to many interesting results. A problem still to solve is to evaluate the likelihood that a sequence will fold into a target structure. This problem can be also viewed as sequence prediction from a known structure.In the current work, we propose improvements in the recent architecture of Geometric Vector Perceptrons in order to optimize the sampling of sequences from a known backbone structure. The proposed model differs from the original in that there is: (i) no updating in the vectorial embedding, only in the scalar one, (ii) only one layer of decoding. The first aspect improves the accuracy of the model and reduces the use of memory, the second allows for training of the model with several tasks without incurring data leakage.We treat the trained classifier as an Energy-Based Model and sample sequences by sampling amino acids in a non-autoreggresive manner in the empty positions of the sequence using energy-guided criteria and followed by Monte Carlo optimization.We improve the median identity of samples from 40.2% to 44.7%.An additional question worth investigating is whether sampled and original sequences fold into similar structures independent of their identity. We chose proteins in our test set whose sampled sequences show low identity (under 30%) but for which our model predicted favorable energies. We used trRosetta server and observed that the predicted structures for sampled sequences highly resemble the predicted structures for original sequences, with an average TM score of 0.848.

L2 Syntactic Alignment in the Reading-Writing Integrated Continuation Task: Evidence From Chinese EFL Learners’ Description of Motion Events

This study investigated how the mode in which the reading-writing integrated continuation task was conducted modulates the effects of second language (L2) syntactic alignment, through the English motion event construction with manner verbs. Ninety Chinese students were assigned to either of the two experimental groups or a control group, and they all experienced a pretest, an alignment phase and a posttest. In the alignment phase, the two experimental groups completed a reading-writing integrated continuation task but in different modes. For the multi-turn mode, participants reconstructed a picture story by continuing the episodes extracted from the story with one episode presented and continued at a time; for the single-turn mode, the first half of the same picture story was presented as a chunk, and then participants read and continued it. Results show that L2 learners aligned with the target structure in completing the story, and the alignment effect was retained in the posttest conducted after a delay of two weeks. Moreover, syntactic alignment was modulated by task mode with the multi-turn group exhibiting stronger immediate and longterm alignment effects. We conclude that the continuation task is a fruitful context for L2 structural alignment, and the magnitude of alignment effect hinges on interactive intensity.

Case Report: Chronic Adaptive Deep Brain Stimulation Personalizing Therapy Based on Parkinsonian State

We describe the case of a 51-year-old man with Parkinson's disease (PD) presenting with motor fluctuations, who received bilateral subthalamic deep brain stimulation (DBS) with an adaptive DBS (aDBS) device, Percept™ PC (Medtronic, Inc. , Minneapolis, MN). This device can deliver electrical stimulations based on fluctuations of neural oscillations of the local field potential (LFP) at the target structure. We observed that the LFP fluctuations were less evident inside the hospital than outside, while the stimulation successfully adapted to beta oscillation fluctuations during the aDBS phase without any stimulation-induced side effects. Thus, this new device facilitates condition-dependent stimulation; this new stimulation method is feasible and provides new insights into the pathophysiological mechanisms of PD.