Search

On the path to full understanding of the structure-function relationship or even design of RNA, structure prediction would offer an intriguing complement to experimental efforts. Any deep learning on RNA structure, however, is hampered by the sparsity of labeled training data. Utilizing the limited data available, we here focus on predicting spatial adjacencies ("contact maps") as a proxy for 3D structure. Our model, BARNACLE, combines the utilization of unlabeled data through self-supervised pre-training and efficient use of the sparse labeled data through an XGBoost classifier. BARNACLE shows a considerable improvement over both the established classical baseline and a deep neural network. In order to demonstrate that our approach can be applied to tasks with similar data constraints, we show that our findings generalize to the related setting of accessible surface area prediction. BARNACLE improves RNA contact prediction with limited labeled data, and such contacts can be used as restraints in RNA structure prediction. [ABSTRACT FROM AUTHOR]

Thermal Bridges on Building Rooftops (TBBR) is a multi-channel remote sensing dataset. It was recorded during six separate UAV fly-overs of the city center of Karlsruhe, Germany, and comprises a total of 926 high-resolution images with 6927 manually-provided thermal bridge annotations. Each image provides five channels: three color, one thermographic, and one computationally derived height map channel. The data is pre-split into training and test data subsets suitable for object detection and instance segmentation tasks. All data is organized and structured to comply with FAIR principles, i.e. being findable, accessible, interoperable, and reusable. It is publicly available and can be downloaded from the Zenodo data repository. This work provides a comprehensive data descriptor for the TBBR dataset to facilitate broad community uptake. [ABSTRACT FROM AUTHOR]

Single-molecule FRET (smFRET) is a versatile technique to study the dynamics and function of biomolecules since it makes nanoscale movements detectable as fluorescence signals. The powerful ability to infer quantitative kinetic information from smFRET data is, however, complicated by experimental limitations. Diverse analysis tools have been developed to overcome these hurdles but a systematic comparison is lacking. Here, we report the results of a blind benchmark study assessing eleven analysis tools used to infer kinetic rate constants from smFRET trajectories. We test them against simulated and experimental data containing the most prominent difficulties encountered in analyzing smFRET experiments: different noise levels, varied model complexity, non-equilibrium dynamics, and kinetic heterogeneity. Our results highlight the current strengths and limitations in inferring kinetic information from smFRET trajectories. In addition, we formulate concrete recommendations and identify key targets for future developments, aimed to advance our understanding of biomolecular dynamics through quantitative experiment-derived models. The ability to infer quantitative kinetic information from single-molecule FRET (smFRET) data can be challenging. Here the authors perform a blind benchmark study assessing different analysis tools used to infer kinetic rate constants from smFRET trajectories, testing on simulated and experimental data. [ABSTRACT FROM AUTHOR]

The spatial organization of chromatin at the scale of topologically associating domains (TADs) and below displays large cell-to-cell variations. Up until now, how this heterogeneity in chromatin conformation is shaped by chromatin condensation, TAD insulation, and transcription has remained mostly elusive. Here, we used Hi-M, a multiplexed DNA-FISH imaging technique providing developmental timing and transcriptional status, to show that the emergence of TADs at the ensemble level partially segregates the conformational space explored by single nuclei during the early development of Drosophila embryos. Surprisingly, a substantial fraction of nuclei display strong insulation even before TADs emerge. Moreover, active transcription within a TAD leads to minor changes to the local inter- and intra-TAD chromatin conformation in single nuclei and only weakly affects insulation to the neighboring TAD. Overall, our results indicate that multiple parameters contribute to shaping the chromatin architecture of single nuclei at the TAD scale. Here the authors applied their recently developed multiplexed DNA-FISH Hi-M method to dissect the sources of heterogeneity in topologically associating domain (TAD)-like organization during Drosophila embryogenesis. This single-nucleus analysis allows them to reveal that multiple parameters contribute to shaping the trace of the chromatin path from a single nucleus. [ABSTRACT FROM AUTHOR]

With increasing data and model complexities, the time required to train neural networks has become prohibitively large. To address the exponential rise in training time, users are turning to data parallel neural networks (DPNN) and large-scale distributed resources on computer clusters. Current DPNN approaches implement the network parameter updates by synchronizing and averaging gradients across all processes with blocking communication operations after each forward-backward pass. This synchronization is the central algorithmic bottleneck. We introduce the distributed asynchronous and selective optimization (DASO) method, which leverages multi-GPU compute node architectures to accelerate network training while maintaining accuracy. DASO uses a hierarchical and asynchronous communication scheme comprised of node-local and global networks while adjusting the global synchronization rate during the learning process. We show that DASO yields a reduction in training time of up to 34% on classical and state-of-the-art networks, as compared to current optimized data parallel training methods. [ABSTRACT FROM AUTHOR]