The ground ice content in cold environments influences the permafrost thermal regime and the thaw trajectories in a warming climate, especially for soils containing excess ice.Despite their importance, the amount and distribution of ground ice are often unknown due to lacking field observations.Hence, modeling the thawing of ice-rich permafrost soi
OTUD5 promotes the growth of hepatocellular carcinoma by deubiquitinating and stabilizing SLC38A1
Abstract Background Deubiquitinating enzymes (DUBs) cleave ubiquitin on substrate molecules to maintain protein stability.DUBs reportedly participate in the tumorigenesis and tumour progression of hepatocellular carcinoma (HCC).OTU deubiquitinase 5 (OTUD5), a DUB family member, has been recognized as a critical regulator in bladder cancer, breast c
Glanidium botocudo, a new species from the rio Doce and rio Mucuri, Minas Gerais, Brazil (Siluriformes: Auchenipteridae) with comments on taxonomic position of Glanidium bockmanni Sarmento-Soares & Buckup
Glanidium botocudo, new species, is described from the tributaries to the upper rio Doce and Mucuri, eastern Minas Gerais State, Brazil.It represents the northernmost record of a centromochlin catfish from the coastal rivers of the Northeastern Atlantic Led Gear Trays Forest.Glanidium botocudo is readily distinguished from its congeners, except Gla
Plasma reforming for enhanced ammonia-air ignition: A numerical study
Using an in-house 0D plasma chemical solver, this paper investigates the species involved in plasma-assisted reforming of both pure ammonia and stoichiometric ammonia-air mixtures.A nanosecond repetitively pulsed plasma is simulated for dielectric barrier discharge conditions, with reduced electric fields of 180 and 360 Td, energies per pulse of 0.
Region-Focusing Data Augmentation via Salient Region Activation and Bitplane Recombination for Target Detection
As the performance of a convolutional neural network is logarithmically proportional to the amount of training data, data augmentation has attracted increasing attention in recent years.Although the current data augmentation methods are efficient because they force the network to learn multiple parts of a given training image through occlusion or r