Tumor suppressor p53 modulates activity-dependent synapse strengthening, autism-like behavior and hippocampus-dependent learning.

Synaptic potentiation underlies various forms of behavior and depends on modulation by multiple activity-dependent transcription factors to coordinate the expression of genes necessary for sustaining synaptic transmission. Our current study identified the tumor suppressor p53 as a novel transcription factor involved in this process. We first revealed that p53 could be elevated upon chemically induced long-term potentiation (cLTP) in cultured primary neurons. By knocking down p53 in neurons, we further showed that p53 is required for cLTP-induced elevation of surface GluA1 and GluA2 subunits of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR). Because LTP is one of the principal plasticity mechanisms underlying behaviors, we employed forebrain-specific knockdown of p53 to evaluate the role of p53 in behavior. Our results showed that, while knocking down p53 in mice does not alter locomotion or anxiety-like behavior, it significantly promotes repetitive behavior and reduces sociability in mice of both sexes. In addition, knocking down p53 also impairs hippocampal LTP and hippocampus-dependent learning and memory. Most importantly, these learning-associated defects are more pronounced in male mice than in female mice, suggesting a sex-specific role of p53 in these behaviors. Using RNA sequencing (RNAseq) to identify p53-associated genes in the hippocampus, we showed that knocking down p53 up- or down-regulates multiple genes with known functions in synaptic plasticity and neurodevelopment. Altogether, our study suggests p53 as an activity-dependent transcription factor that mediates the surface expression of AMPAR, permits hippocampal synaptic plasticity, represses autism-like behavior, and promotes hippocampus-dependent learning and memory.

Preoperative serum level of CA153 and a new model to predict the sub-optimal primary debulking surgery in patients with advanced epithelial ovarian cancer.

OBJECTIVE: The aim of this study was to establish a preoperative model to predict the outcome of primary debulking surgery (PDS) for advanced ovarian cancer (AOC) patients by combing Suidan predictive model with HE4, CA125, CA153 and ROMA index. METHODS: 76 AOC Patients in revised 2014 International Federation of Gynecology and Obstetrics (FIGO) stage III-IV who underwent PDS between 2017 and 2019 from Yunnan Cancer Hospital were included. Clinical data including the levels of preoperative serum HE4, CA125, CA153 and mid-lower abdominal CT-enhanced scan results were collected. The logistics regression analysis was performed to find factors associated with sub-optimal debulking surgery (SDS). The receiver operating characteristic curve was used to evaluate the predictive performances of selected variables in the outcome of primary debulking surgery. The predictive index value (PIV) model was constructed to predict the outcome of SDS. RESULTS: Optimal surgical cytoreduction was achieved in 61.84% (47/76) patients. The value for CA125, HE4, CA153, ROMA index and Suidan score was lower in optimal debulking surgery (ODS) group than SDS group. Based on the Youden index, which is widely used for evaluating the performance of predictive models, the best cutoff point for the preoperative serum HE4, CA125, CA153, ROMA index and Suidan score to distinguish SDS were 431.55 pmol/l, 2277 KU/L, 57.19 KU/L, 97.525% and 2.5, respectively. Patients with PIV≥5 may not be able to achieve optimal surgical cytoreduction. The diagnostic accuracy, NPV, PPV and specificity for diagnosing SDS were 73.7%, 82.9%, 62.9% and 72.3%, respectively. In the constructed model, the AUC of the SDS prediction was 0.770 (95% confidence interval: 0.654-0.887), P<0.001. CONCLUSION: Preoperative serum CA153 level is an important non-invasive predictor of primary SDS in advanced AOC, which has not been reported before. The constructed PIV model based on Suidan's predictive model plus HE4, CA125, CA153 and ROMA index can noninvasively predict SDS in AOC patients, the accuracy of this prediction model still needs to be validated in future studies.

Intrinsic ferroelectricity in Y-doped HfO2 thin films.

Ferroelectric HfO2-based materials hold great potential for the widespread integration of ferroelectricity into modern electronics due to their compatibility with existing Si technology. Earlier work indicated that a nanometre grain size was crucial for the stabilization of the ferroelectric phase. This constraint, associated with a high density of structural defects, obscures an insight into the intrinsic ferroelectricity of HfO2-based materials. Here we demonstrate that stable and enhanced polarization can be achieved in epitaxial HfO2 films with a high degree of structural order (crystallinity). An out-of-plane polarization value of 50 µC cm-2 has been observed at room temperature in Y-doped HfO2(111) epitaxial thin films, with an estimated full value of intrinsic polarization of 64 µC cm-2, which is in close agreement with density functional theory calculations. The crystal structure of films reveals the Pca21 orthorhombic phase with small rhombohedral distortion, underlining the role of the structural constraint in stabilizing the ferroelectric phase. Our results suggest that it could be possible to exploit the intrinsic ferroelectricity of HfO2-based materials, optimizing their performance in device applications.

Enhancingß-Ga2O3-film ultraviolet detectors via RF magnetron sputtering with seed layer insertion onc-plane sapphire substrate.

Gallium oxide (Ga2O3) possesses a band gap of approximately 4.9 eV, aligning its detection wavelength within the solar-blind region, making it an ideal semiconductor material for solar-blind photodetectors. This study aims to enhance the performance of Ga2O3ultraviolet (UV) detectors by pre-depositing a Ga2O3seed layer on ac-plane sapphire substrate. The x-ray diffraction and x-ray photoelectron spectroscopy analyses validated that the deposited films, following high-temperature annealing, comprisedß-Ga2O3. Comparing samples with and without a 20 nm seed layer, it was found that the former exhibited fewer oxygen defects and substantially improved crystal quality. The incorporation of the seed layer led to the realization of detectors with remarkably low dark current (≤15.3 fA). Moreover, the photo-to-dark current ratio was enhanced by 30% (surpassing 1.3 × 104) and the response/recovery time reduced to 0.9 s/0.01 s, indicating faster performance. Furthermore, these detectors demonstrated higher responsivity (4.8 mA W-1), improved detectivity (2.49 × 1016Jones), and excellent solar-blind characteristics. This study serves as a foundational stepping toward achieving high-qualityß-Ga2O3thin film and UV detector arrays.

Active learning to classify macromolecular structures in situ for less supervision in cryo-electron tomography.

MOTIVATION: Cryo-Electron Tomography (cryo-ET) is a 3D bioimaging tool that visualizes the structural and spatial organization of macromolecules at a near-native state in single cells, which has broad applications in life science. However, the systematic structural recognition and recovery of macromolecules captured by cryo-ET are difficult due to high structural complexity and imaging limits. Deep learning-based subtomogram classification has played critical roles for such tasks. As supervised approaches, however, their performance relies on sufficient and laborious annotation on a large training dataset. RESULTS: To alleviate this major labeling burden, we proposed a Hybrid Active Learning (HAL) framework for querying subtomograms for labeling from a large unlabeled subtomogram pool. Firstly, HAL adopts uncertainty sampling to select the subtomograms that have the most uncertain predictions. This strategy enforces the model to be aware of the inductive bias during classification and subtomogram selection, which satisfies the discriminativeness principle in AL literature. Moreover, to mitigate the sampling bias caused by such strategy, a discriminator is introduced to judge if a certain subtomogram is labeled or unlabeled and subsequently the model queries the subtomogram that have higher probabilities to be unlabeled. Such query strategy encourages to match the data distribution between the labeled and unlabeled subtomogram samples, which essentially encodes the representativeness criterion into the subtomogram selection process. Additionally, HAL introduces a subset sampling strategy to improve the diversity of the query set, so that the information overlap is decreased between the queried batches and the algorithmic efficiency is improved. Our experiments on subtomogram classification tasks using both simulated and real data demonstrate that we can achieve comparable testing performance (on average only 3% accuracy drop) by using less than 30% of the labeled subtomograms, which shows a very promising result for subtomogram classification task with limited labeling resources. AVAILABILITY AND IMPLEMENTATION: https://github.com/xulabs/aitom. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Coupled mixed model for joint genetic analysis of complex disorders with two independently collected data sets.

BACKGROUND: In the last decade, Genome-wide Association studies (GWASs) have contributed to decoding the human genome by uncovering many genetic variations associated with various diseases. Many follow-up investigations involve joint analysis of multiple independently generated GWAS data sets. While most of the computational approaches developed for joint analysis are based on summary statistics, the joint analysis based on individual-level data with consideration of confounding factors remains to be a challenge. RESULTS: In this study, we propose a method, called Coupled Mixed Model (CMM), that enables a joint GWAS analysis on two independently collected sets of GWAS data with different phenotypes. The CMM method does not require the data sets to have the same phenotypes as it aims to infer the unknown phenotypes using a set of multivariate sparse mixed models. Moreover, CMM addresses the confounding variables due to population stratification, family structures, and cryptic relatedness, as well as those arising during data collection such as batch effects that frequently appear in joint genetic studies. We evaluate the performance of CMM using simulation experiments. In real data analysis, we illustrate the utility of CMM by an application to evaluating common genetic associations for Alzheimer's disease and substance use disorder using datasets independently collected for the two complex human disorders. Comparison of the results with those from previous experiments and analyses supports the utility of our method and provides new insights into the diseases. The software is available at https://github.com/HaohanWang/CMM .

Nitride-Oxide-Metal Heterostructure with Self-Assembled Core-Shell Nanopillar Arrays: Effect of Ordering on Magneto-Optical Properties.

Magneto-optical (MO) coupling incorporates photon-induced change of magnetic polarization that can be adopted in ultrafast switching, optical isolators, mode convertors, and optical data storage components for advanced optical integrated circuits. However, integrating plasmonic, magnetic, and dielectric properties in one single material system poses challenges since one natural material can hardly possess all these functionalities. Here, co-deposition of a three-phase heterostructure composed of a durable conductive nitride matrix with embedded core-shell vertically aligned nanopillars, is demonstrated. The unique coupling between ferromagnetic NiO core and atomically sharp plasmonic Au shell enables strong MO activity out-of-plane at room temperature. Further, a template growth process is applied, which significantly enhances the ordering of the nanopillar array. The ordered nanostructure offers two schemes of spin polarization which result in stronger antisymmetry of Kerr rotation. The presented complex hybrid metamaterial platform with strong magnetic and optical anisotropies is promising for tunable and modulated all-optical-based nanodevices.

Poly(A)-DG: A deep-learning-based domain generalization method to identify cross-species Poly(A) signal without prior knowledge from target species.

In eukaryotes, polyadenylation (poly(A)) is an essential process during mRNA maturation. Identifying the cis-determinants of poly(A) signal (PAS) on the DNA sequence is the key to understand the mechanism of translation regulation and mRNA metabolism. Although machine learning methods were widely used in computationally identifying PAS, the need for tremendous amounts of annotation data hinder applications of existing methods in species without experimental data on PAS. Therefore, cross-species PAS identification, which enables the possibility to predict PAS from untrained species, naturally becomes a promising direction. In our works, we propose a novel deep learning method named Poly(A)-DG for cross-species PAS identification. Poly(A)-DG consists of a Convolution Neural Network-Multilayer Perceptron (CNN-MLP) network and a domain generalization technique. It learns PAS patterns from the training species and identifies PAS in target species without re-training. To test our method, we use four species and build cross-species training sets with two of them and evaluate the performance of the remaining ones. Moreover, we test our method against insufficient data and imbalanced data issues and demonstrate that Poly(A)-DG not only outperforms state-of-the-art methods but also maintains relatively high accuracy when it comes to a smaller or imbalanced training set.

Metal-Free Oxide-Nitride Heterostructure as a Tunable Hyperbolic Metamaterial Platform.

Metal-free plasmonic metamaterials with wide-range tunable optical properties are highly desired for various components in future integrated optical devices. Designing a ceramic-ceramic hybrid metamaterial has been theoretically proposed as a solution to this critical optical material demand. However, the processing of such all-ceramic metamaterials is challenging due to difficulties in integrating two very dissimilar ceramic phases as one hybrid system. In this work, an oxide-nitride hybrid metamaterial combining two highly dissimilar ceramic phases, i.e., semiconducting weak ferromagnetic NiO nanorods and conductive plasmonic TiN matrix, has been successfully integrated as a unique vertically aligned nanocomposite form. Highly anisotropic optical properties such as hyperbolic dispersions and strong magneto-optical coupling have been demonstrated under room temperature. The novel functionalities presented show the strong potentials of this new ceramic-ceramic hybrid thin film platform and its future applications in next-generation nanophotonics and magneto-optical integrated devices without the lossy metallic components.

Precision Lasso: accounting for correlations and linear dependencies in high-dimensional genomic data.

MOTIVATION: Association studies to discover links between genetic markers and phenotypes are central to bioinformatics. Methods of regularized regression, such as variants of the Lasso, are popular for this task. Despite the good predictive performance of these methods in the average case, they suffer from unstable selections of correlated variables and inconsistent selections of linearly dependent variables. Unfortunately, as we demonstrate empirically, such problematic situations of correlated and linearly dependent variables often exist in genomic datasets and lead to under-performance of classical methods of variable selection. RESULTS: To address these challenges, we propose the Precision Lasso. Precision Lasso is a Lasso variant that promotes sparse variable selection by regularization governed by the covariance and inverse covariance matrices of explanatory variables. We illustrate its capacity for stable and consistent variable selection in simulated data with highly correlated and linearly dependent variables. We then demonstrate the effectiveness of the Precision Lasso to select meaningful variables from transcriptomic profiles of breast cancer patients. Our results indicate that in settings with correlated and linearly dependent variables, the Precision Lasso outperforms popular methods of variable selection such as the Lasso, the Elastic Net and Minimax Concave Penalty (MCP) regression. AVAILABILITY AND IMPLEMENTATION: Software is available at https://github.com/HaohanWang/thePrecisionLasso. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Deep mixed model for marginal epistasis detection and population stratification correction in genome-wide association studies.

BACKGROUND: Genome-wide Association Studies (GWAS) have contributed to unraveling associations between genetic variants in the human genome and complex traits for more than a decade. While many works have been invented as follow-ups to detect interactions between SNPs, epistasis are still yet to be modeled and discovered more thoroughly. RESULTS: In this paper, following the previous study of detecting marginal epistasis signals, and motivated by the universal approximation power of deep learning, we propose a neural network method that can potentially model arbitrary interactions between SNPs in genetic association studies as an extension to the mixed models in correcting confounding factors. Our method, namely Deep Mixed Model, consists of two components: 1) a confounding factor correction component, which is a large-kernel convolution neural network that focuses on calibrating the residual phenotypes by removing factors such as population stratification, and 2) a fixed-effect estimation component, which mainly consists of an Long-short Term Memory (LSTM) model that estimates the association effect size of SNPs with the residual phenotype. CONCLUSIONS: After validating the performance of our method using simulation experiments, we further apply it to Alzheimer's disease data sets. Our results help gain some explorative understandings of the genetic architecture of Alzheimer's disease.

Variable selection in heterogeneous datasets: A truncated-rank sparse linear mixed model with applications to genome-wide association studies.

A fundamental and important challenge in modern datasets of ever increasing dimensionality is variable selection, which has taken on renewed interest recently due to the growth of biological and medical datasets with complex, non-i.i.d. structures. Naïvely applying classical variable selection methods such as the Lasso to such datasets may lead to a large number of false discoveries. Motivated by genome-wide association studies in genetics, we study the problem of variable selection for datasets arising from multiple subpopulations, when this underlying population structure is unknown to the researcher. We propose a unified framework for sparse variable selection that adaptively corrects for population structure via a low-rank linear mixed model. Most importantly, the proposed method does not require prior knowledge of sample structure in the data and adaptively selects a covariance structure of the correct complexity. Through extensive experiments, we illustrate the effectiveness of this framework over existing methods. Further, we test our method on three different genomic datasets from plants, mice, and human, and discuss the knowledge we discover with our method.

Multiplex confounding factor correction for genomic association mapping with squared sparse linear mixed model.

Genome-wide Association Study has presented a promising way to understand the association between human genomes and complex traits. Many simple polymorphic loci have been shown to explain a significant fraction of phenotypic variability. However, challenges remain in the non-triviality of explaining complex traits associated with multifactorial genetic loci, especially considering the confounding factors caused by population structure, family structure, and cryptic relatedness. In this paper, we propose a Squared-LMM (LMM2) model, aiming to jointly correct population and genetic confounding factors. We offer two strategies of utilizing LMM2 for association mapping: 1) It serves as an extension of univariate LMM, which could effectively correct population structure, but consider each SNP in isolation. 2) It is integrated with the multivariate regression model to discover association relationship between complex traits and multifactorial genetic loci. We refer to this second model as sparse Squared-LMM (sLMM2). Further, we extend LMM2/sLMM2 by raising the power of our squared model to the LMMn/sLMMn model. We demonstrate the practical use of our model with synthetic phenotypic variants generated from genetic loci of Arabidopsis Thaliana. The experiment shows that our method achieves a more accurate and significant prediction on the association relationship between traits and loci. We also evaluate our models on collected phenotypes and genotypes with the number of candidate genes that the models could discover. The results suggest the potential and promising usage of our method in genome-wide association studies.

Tuning the Néel Temperature of Hexagonal Ferrites by Structural Distortion.

To tune the magnetic properties of hexagonal ferrites, a family of magnetoelectric multiferroic materials, by atomic-scale structural engineering, we studied the effect of structural distortion on the magnetic ordering temperature (T_{N}) in these materials. Using the symmetry analysis, we show that unlike most antiferromagnetic rare-earth transition-metal perovskites, a larger structural distortion leads to a higher T_{N} in hexagonal ferrites and manganites, because the K_{3} structural distortion induces the three-dimensional magnetic ordering, which is forbidden in the undistorted structure by symmetry. We also revealed a near-linear relation between T_{N} and the tolerance factor and a power-law relation between T_{N} and the K_{3} distortion amplitude. Following the analysis, a record-high T_{N} (185 K) among hexagonal ferrites was predicted in hexagonal ScFeO_{3} and experimentally verified in epitaxially stabilized films. These results add to the paradigm of spin-lattice coupling in antiferromagnetic oxides and suggests further tunability of hexagonal ferrites if more lattice distortion can be achieved.

Backward genotype-transcript-phenotype association mapping.

Genome-wide association studies have discovered a large number of genetic variants associated with complex diseases such as Alzheimer's disease. However, the genetic background of such diseases is largely unknown due to the complex mechanisms underlying genetic effects on traits, as well as a small sample size (e.g., 1000) and a large number of genetic variants (e.g., 1 million). Fortunately, datasets that contain genotypes, transcripts, and phenotypes are becoming more readily available, creating new opportunities for detecting disease-associated genetic variants. In this paper, we present a novel approach called "Backward Three-way Association Mapping" (BTAM) for detecting three-way associations among genotypes, transcripts, and phenotypes. Assuming that genotypes affect transcript levels, which in turn affect phenotypes, we first find transcripts associated with the phenotypes, and then find genotypes associated with the chosen transcripts. The backward ordering of association mappings allows us to avoid a large number of association testings between all genotypes and all transcripts, making it possible to identify three-way associations with a small computational cost. In our simulation study, we demonstrate that BTAM significantly improves the statistical power over "forward" three-way association mapping that finds genotypes associated with both transcripts and phenotypes and genotype-phenotype association mapping. Furthermore, we apply BTAM on an Alzheimer's disease dataset and report top 10 genotype-transcript-phenotype associations.

Six1 induces protein synthesis signaling expression in duck myoblasts mainly via up-regulation of mTOR.

As a critical transcription factor, Six1 plays an important role in the regulation of myogenesis and muscle development. However, little is known about its regulatory mechanism associated with muscular protein synthesis. The objective of this study was to investigate the effects of overexpression ofSix1 on the expression of key protein metabolism-related genes in duck myoblasts. Through an experimental model where duck myoblasts were transfected with a pEGFP-duSix1 construct, we found that overexpression of duckSix1 could enhance cell proliferation activity and increase mRNA expression levels of key genes involved in the PI3K/Akt/mTOR signaling pathway, while the expression of FOXO1, MuRF1and MAFbx was not significantly altered, indicating thatSix1 could promote protein synthesis in myoblasts through up-regulating the expression of several related genes. Additionally, in duck myoblasts treated with LY294002 and rapamycin, the specific inhibitors ofPI3K and mTOR, respectively, the overexpression of Six1 could significantly ameliorate inhibitive effects of these inhibitors on protein synthesis. Especially, the mRNA expression levels of mTOR and S6K1 were observed to undergo a visible change, and a significant increase in protein expression of S6K1 was seen. These data suggested that Six1plays an important role in protein synthesis, which may be mainly due to activation of the mTOR signaling pathway.

Impact of thermal stress during incubation on gene expression in embryonic muscle of Peking ducks (Anasplatyrhynchos domestica).

Changes in temperature will influence poultry embryonic muscle development. However, little is known about the changes in molecular processes impacted by incubation temperature in avians. In this study, we investigated the effects of increasing the incubation temperature by 1°C from day 11-20 on the embryonic and posthatch skeletal muscle development of the Peking duck, and identified the differentially expressed genes using RNA-seq of leg muscle tissues. The results showed that altering the incubation temperature had immediate and long-lasting effects on phenotypic changes in the embryonic and post-hatching muscle development. It was shown that expression levels of total 1370 genes were altered in muscle tissues by the thermal treatments. The gene ontology (GO) analyses indicated that cellular processes including metabolism, cell cycle, catalytic activity, and enzyme regulatory activity may have involved in the muscle mass impacted by thermal manipulation. TGF-beta and insulin pathways as two classical muscle development related pathways may also involve in regulating muscle mass. These data may be helpful for understanding the physiological and biochemical processes of muscle development under environmental treatments in embryonic avians.

Silencing Pax3 by shRNA inhibits the proliferation and differentiation of duck (Anas platyrhynchos) myoblasts.

The Pax3 gene has been proven to play a crucial role in determining myogenic progenitor cell fate during embryonic myogenesis; however, the molecular role of Pax3 in myoblast development during later stages of myogenesis is unknown. We hypothesized that Pax3 would function in myoblast proliferation and differentiation; therefore, we employed three short hairpin RNAs (shRNAs) (shRNA1, shRNA2, and shRNA3) that target Pax3 to characterize the function of Pax3 in duck myoblast development. The mRNA and protein expression levels of Pax3 in duck myoblasts were detected using real-time PCR and Western blotting. Cell proliferation was assessed using the MTT and BrdU assays, while cell differentiation was assayed using immunofluorescence labeling with a MyoG antibody. Additionally, folic acid (FA), which is a rescue tool, was added into the medium of duck myoblasts to indirectly examine the function of Pax3 on duck myoblast proliferation and differentiation. The results revealed that one of the shRNA vectors, shRNA1, could significantly and stably reduce the expression of Pax3 (P < 0.05). Silencing Pax3 by shRNA1 significantly reduced the proliferation and differentiation of duck myoblasts (P < 0.05) due to downregulated expression of myogenic regulator factors. These trends could be rescued by adding FA; and Pax7, a paralog gene of Pax3, was involved in those processes. Overall, Pax3 had a positive function in duck myoblast proliferation and differentiation by modulating the expression of myogenic regulation factors, and shRNA targeting of Pax3 might be a new approach for understanding the function of Pax3 in the development of diverse tissues.

Molecular cloning and expression pattern of duck Six1 and its preliminary functional analysis in myoblasts transfected with eukaryotic expression vector.

Skeletal muscle development is regulated by Six1, an important myogenic transcription factor. However, the functional analysis of duck Six1 has not been reported. Here, we cloned the coding domain sequence (CDS) region of the duck Six1 gene using RT-PCR and RACE methods. Bioinformatics analysis revealed that duck Six1 CDS region comprised of 849 bp and encoded 282 amino acids and had a high degree of homology with other species, suggesting that the functions of duck Six1 gene are conserved among other animals. Real-time PCR used to determine the mRNA expression profiles of duck Six1 in different tissues and different developmental stages showed that Six1 was highly expressed in skeletal muscle and the embryonic stage. Furthermore, the eukaryotic expression vector pEGFP-duSix1 was constructed and transfected into the duck myoblasts; the MTT assay revealed an obvious increase of cell proliferation after transfection. The expression profiles of Six1, Myf5 and MyoD showed that their expression levels were significantly increased. These results together suggested that pEGFP-duSix1 vector was constructed successfully and overexpression of duck Six1 in the myoblasts could promote cell proliferation activity and significant up-regulate expression of Myf5 and MyoD.

BadLabel: A Robust Perspective on Evaluating and Enhancing Label-Noise Learning.

Label-noise learning (LNL) aims to increase the model's generalization given training data with noisy labels. To facilitate practical LNL algorithms, researchers have proposed different label noise types, ranging from class-conditional to instance-dependent noises. In this paper, we introduce a novel label noise type called BadLabel, which can significantly degrade the performance of existing LNL algorithms by a large margin. BadLabel is crafted based on the label-flipping attack against standard classification, where specific samples are selected and their labels are flipped to other labels so that the loss values of clean and noisy labels become indistinguishable. To address the challenge posed by BadLabel, we further propose a robust LNL method that perturbs the labels in an adversarial manner at each epoch to make the loss values of clean and noisy labels again distinguishable. Once we select a small set of (mostly) clean labeled data, we can apply the techniques of semi-supervised learning to train the model accurately. Empirically, our experimental results demonstrate that existing LNL algorithms are vulnerable to the newly introduced BadLabel noise type, while our proposed robust LNL method can effectively improve the generalization performance of the model under various types of label noise. The new dataset of noisy labels and the source codes of robust LNL algorithms are available at https://github.com/zjfheart/BadLabels.