Browse Articles
Discover research articles across all indexed journals
Strengthening power grid projects’ governance and sustainability through lifecycle auditing
Exploring colostrum microbiota and its influence on early calf gut microbiota development using full-length 16S rRNA gene metabarcoding
Research on conveyor belt damage detection method based on FDEP−YOLOv8
Lead micro- and nanoparticles directly observed within gunshot wounds in hunted game meat
A generalized framework for the collinear restricted four-body problem with a central dominant mass
Balance control in children and adolescents with intellectual disability: a systematic review and meta-analysis
Risk assessment of pARDS in severe pneumonia patients based on lung injury prediction scores and serum biomarkers
Experimental study on vertical compressive bearing capacity of single pile in Weihe river second terrace
Motor imagery in individuals with congenital aphantasia
Explicit error coding can mediate gain recalibration in continuous bump attractor networks
Integrating spatial and behavioral data provides comprehensive assessment of grizzly bear-ecotourism coexistence in Nuxalk Territory
Dynamic identification of reactive iron-oxo species in heterogeneous fenton-like reaction via operando stopped-flow IR spectroscopy
Analysis of fractional-order model for the transmission dynamics of malaria via Caputo–Fabrizio and Atangana–Baleanu operators
Southern Ocean influence on Atlantic Meridional Overturning Circulation across climate states
Accelerating the tuning process for optimizing DNN operators by ROFT model
Abstract Deep neural networks (DNNs) are computationally intensive and optimized in different ways. Some compiler optimizations for DNNs could achieve performance almost the same as, or even better than, manual optimizations. However, the former mechanisms usually require an unbearably long optimization time in the tuning process. In this paper, we propose a new method that accelerates the tuning process significantly without performance penalties. In particular, we use a Roofline-like cost model, namely ROFT (Roofline for Fast AutoTune), to evaluate the performance of schedules. The ROFT model can be easily implemented on different microarchitectures, e.g., NVidia GPUs and Huawei Ascend NPUs. Based on the cost model, we implement a flexible two-stage search algorithm, which significantly improves the time of tuning process. Experiments show that the ROFT method speeds up the tuning process by about 4X and 10X compared with AutoTVM on NVidia GPUs and the AutoTune of Huawei’s Tensor Boost Engine (TBE) on Huawei Ascend310 NPUs for some typical DNNs, respectively. It improves the inference time of some DNNs by up to 7% as well.
Author Correction: Cross-species comparison reveals therapeutic vulnerabilities halting glioblastoma progression
NR4A1 expression aberrations contribute to radiotherapy resistance in gastric cancer
Abstract Gastric cancer (GC) is a common and life-threatening malignancy. It is often diagnosed at advanced stages. A major challenge in treatment is the development of radiotherapy resistance. This resistance significantly reduces the effectiveness of therapy. Therefore, our objective is to pinpoint genes linked to the resistance against radiotherapy in gastric cancer. The NR4A1 expression exhibited abnormalities in radioresistant gastric cancer cells, specifically MKN45-R and AGS-R cells, in comparison to their respective parental cells. Through high-throughput sequencing, we screened these cells and utilized quantitative reverse transcription-polymerase chain reaction analysis (qRT-PCR) to quantify NR4A1 expression. Functional analyses of NR4A1 in GC radioresistance were conducted by overexpressing NR4A1 in both MKN45-R and AGS-R cells, as well as their parental counterparts. Finally, the clinical relevance of NR4A1 in GC was confirmed through animal experiments. The expression of NR4A1 was significantly reduced in radioresistant cells, namely MKN45-R and AGS-R, as opposed to their parental counterparts. Overexpressing NR4A1 in MKN45-R and AGS-R cells resulted in heightened sensitivity to radiation. Additionally, NR4A1 overexpression exerted inhibitory effects on the growth and metastasis of GC tissues in animal models following irradiation (IR). Our observations indicate that NR4A1 plays a crucial role in enhancing the radiation sensitivity of gastric cancer cells, underscoring the therapeutic potential of NR4A1 in GC treatment.