A single stage knowledge distillation network for brain tumor segmentation on limited MR image modalities.

BACKGROUND AND OBJECTIVE: Precisely segmenting brain tumors using multimodal Magnetic Resonance Imaging (MRI) is an essential task for early diagnosis, disease monitoring, and surgical planning. Unfortunately, the complete four image modalities utilized in the well-known BraTS benchmark dataset: T1, T2, Fluid-Attenuated Inversion Recovery (FLAIR), and T1 Contrast-Enhanced (T1CE) are not regularly acquired in clinical practice due to the high cost and long acquisition time. Rather, it is common to utilize limited image modalities for brain tumor segmentation. METHODS: In this paper, we propose a single stage learning of knowledge distillation algorithm that derives information from the missing modalities for better segmentation of brain tumors. Unlike the previous works that adopted a two-stage framework to distill the knowledge from a pre-trained network into a student network, where the latter network is trained on limited image modality, we train both models simultaneously using a single-stage knowledge distillation algorithm. We transfer the information by reducing the redundancy from a teacher network trained on full image modalities to the student network using Barlow Twins loss on a latent-space level. To distill the knowledge on the pixel level, we further employ a deep supervision idea that trains the backbone networks of both teacher and student paths using Cross-Entropy loss. RESULTS: We demonstrate that the proposed single-stage knowledge distillation approach enables improving the performance of the student network in each tumor category with overall dice scores of 91.11% for Tumor Core, 89.70% for Enhancing Tumor, and 92.20% for Whole Tumor in the case of only using the FLAIR and T1CE images, outperforming the state-of-the-art segmentation methods. CONCLUSIONS: The outcomes of this work prove the feasibility of exploiting the knowledge distillation in segmenting brain tumors using limited image modalities and hence make it closer to clinical practices.

TRIP13 Participates in Immediate-Early Sensing of DNA Strand Breaks and ATM Signaling Amplification through MRE11.

Thyroid hormone receptor-interacting protein 13 (TRIP13) participates in various regulatory steps related to the cell cycle, such as the mitotic spindle assembly checkpoint and meiotic recombination, possibly by interacting with members of the HORMA domain protein family. Recently, it was reported that TRIP13 could regulate the choice of the DNA repair pathway, i.e., homologous recombination (HR) or nonhomologous end-joining (NHEJ). However, TRIP13 is recruited to DNA damage sites within a few seconds after damage and may therefore have another function in DNA repair other than regulation of the pathway choice. Furthermore, the depletion of TRIP13 inhibited both HR and NHEJ, suggesting that TRIP13 plays other roles besides regulation of choice between HR and NHEJ. To explore the unidentified functions of TRIP13 in the DNA damage response, we investigated its genome-wide interaction partners in the context of DNA damage using quantitative proteomics with proximity labeling. We identified MRE11 as a novel interacting partner of TRIP13. TRIP13 controlled the recruitment of MDC1 to DNA damage sites by regulating the interaction between MDC1 and the MRN complex. Consistently, TRIP13 was involved in ATM signaling amplification. Our study provides new insight into the function of TRIP13 in immediate-early DNA damage sensing and ATM signaling activation.

3DMesh-GAR: 3D Human Body Mesh-Based Method for Group Activity Recognition.

Group activity recognition is a prime research topic in video understanding and has many practical applications, such as crowd behavior monitoring, video surveillance, etc. To understand the multi-person/group action, the model should not only identify the individual person's action in the context but also describe their collective activity. A lot of previous works adopt skeleton-based approaches with graph convolutional networks for group activity recognition. However, these approaches are subject to limitation in scalability, robustness, and interoperability. In this paper, we propose 3DMesh-GAR, a novel approach to 3D human body Mesh-based Group Activity Recognition, which relies on a body center heatmap, camera map, and mesh parameter map instead of the complex and noisy 3D skeleton of each person of the input frames. We adopt a 3D mesh creation method, which is conceptually simple, single-stage, and bounding box free, and is able to handle highly occluded and multi-person scenes without any additional computational cost. We implement 3DMesh-GAR on a standard group activity dataset: the Collective Activity Dataset, and achieve state-of-the-art performance for group activity recognition.

Ewing sarcoma protein promotes dissociation of poly(ADP-ribose) polymerase 1 from chromatin.

Poly(ADP-ribose) polymerase 1 (PARP1) facilitates DNA damage response (DDR). While the Ewing's sarcoma breakpoint region 1 (EWS) protein fused to FLI1 triggers sarcoma formation, the physiological function of EWS is largely unknown. Here, we investigate the physiological role of EWS in regulating PARP1. We show that EWS is required for PARP1 dissociation from damaged DNA. Abnormal PARP1 accumulation caused by EWS inactivation leads to excessive Poly(ADP-Ribosy)lation (PARylation) and triggers cell death in both in vitro and in vivo models. Consistent with previous work, the arginine-glycine-glycine (RGG) domain of EWS is essential for PAR chain interaction and PARP1 dissociation from damaged DNA. Ews and Parp1 double mutant mice do not show improved survival, but supplementation with nicotinamide mononucleotides extends Ews-mutant pups' survival, which might be due to compensatory activation of other PARP proteins. Consistently, PARP1 accumulates on chromatin in Ewing's sarcoma cells expressing an EWS fusion protein that cannot interact with PARP1, and tissues derived from Ewing's sarcoma patients show increased PARylation. Taken together, our data reveal that EWS is important for removing PARP1 from damaged chromatin.

A Deep Learning-Based Approach for Identifying the Medicinal Uses of Plant-Derived Natural Compounds.

Medicinal plants and their extracts have been used as important sources for drug discovery. In particular, plant-derived natural compounds, including phytochemicals, antioxidants, vitamins, and minerals, are gaining attention as they promote health and prevent disease. Although several in vitro methods have been developed to confirm the biological activities of natural compounds, there is still considerable room to reduce time and cost. To overcome these limitations, several in silico methods have been proposed for conducting large-scale analysis, but they are still limited in terms of dealing with incomplete and heterogeneous natural compound data. Here, we propose a deep learning-based approach to identify the medicinal uses of natural compounds by exploiting massive and heterogeneous drug and natural compound data. The rationale behind this approach is that deep learning can effectively utilize heterogeneous features to alleviate incomplete information. Based on latent knowledge, molecular interactions, and chemical property features, we generated 686 dimensional features for 4,507 natural compounds and 2,882 approved and investigational drugs. The deep learning model was trained using the generated features and verified drug indication information. When the features of natural compounds were applied as input to the trained model, potential efficacies were successfully predicted with high accuracy, sensitivity, and specificity.

SHPRH as a new player in ribosomal RNA transcription and its potential role in homeostasis of ribosomal DNA repeats.

There are hundreds of copies of rDNA repeats in mammalian chromosomes and the ratio of active, poised, or inactive rDNA is regulated in epigenetic manners. Recent studies demonstrated that a post-DNA replication repair enzyme, SHPRH affects rRNA transcription by recognizing epigenetic markers on rDNA promoters and unveiled potential links between DNA repair and ribosome biogenesis. This study suggests that SHPRH could be a link between mTOR-mediated epigenetic regulations and rRNA transcription, while concomitantly affecting genomic integrity.