With the rapid advancement of deep neural networks (DNNs), model watermarking has emerged as a widely adopted technique for safeguarding model copyrights. A prevalent method involves utilizing a watermark decoder to retrieve watermark bits from generated outputs, but such methods are often vulnerable to model fine-tuning attacks. Traditionally, this challenge is mitigated through adversarial training or data augmentation, both of which significantly increase the computational burden. In this paper, we present a solution employing Flat Minima Aware (FMA) optimizers to bolster the robustness of model watermarking without requiring additional training data. By optimizing the watermark loss with flat minima awareness, our approaches significantly enhance the robustness of watermarks against the model fine-tuning attack. Comprehensive experiments have demonstrated our method’s superior ability to preserve watermark integrity. These findings suggest that this innovative optimization strategy offers a robust and efficient pathway for protecting models, thereby contributing to more secure and reliable model copyright protection mechanisms.

Due to the widespread use and proliferation of Deep Neural Networks (DNNs), safeguarding their Intellectual Property Rights (IPR) has become increasingly important. This article proposes a method for watermarking a cyclic Generative Adversarial Network (GAN), specifically CycleGAN, to address the gap between the watermarking of conventional GAN models and cyclic GAN watermarking. The proposed method involves training a watermark decoder, which is then frozen and used to extract the watermark bits during the training of the CycleGAN model. The model is trained using specific loss functions that are optimized to achieve excellent performance on both the Image-to-Image Translation (I2IT) task and watermark embedding. Besides, a comprehensive theoretical and practical statistical analysis to verify the ownership of the model from the extracted watermark bits is given. At last, the model's robustness is evaluated against image post-processing, and further improved by fine-tuning the watermark decoder by applying data augmentation to the generated images before extracting the watermark bits. We also verify the robustness of the watermark to surrogate model attacks, carried out by accessing the watermarked model in a black-box modality. The experimental results demonstrate that the proposed method is effective and robust against image post-processing and can resist surrogate model attacks.

Amid the burgeoning development of generative models like diffusion models, the task of differentiating synthesized audio from its natural counterpart grows more daunting. Deepfake detection offers a viable solution to combat this challenge. Yet, this defensive measure unintentionally fuels the continued refinement of generative models. Watermarking emerges as a proactive and sustainable tactic, preemptively regulating the creation and dissemination of synthesized content. Thus, this paper, as a pioneer, proposes the generative robust audiowatermarking method (Groot), presenting a paradigm for proactively supervising the synthesized audio and its source diffusion models. In this paradigm, the processes of watermark generation and audio synthesis occur simultaneously, facilitated by parameter-fixed diffusion models equipped with a dedicated encoder. The watermark embedded within the audio can subsequently be retrieved by a lightweight decoder. The experimental results highlight Groot's outstanding performance, particularly in terms of robustness, surpassing that of the leading state-of-the-art methods. Beyond its impressive resilience against individual post-processing attacks, Groot exhibits exceptional robustness when facing compound attacks, maintaining an average watermark extraction accuracy of around 95%. Our audio samples are available at https://groot-gaw.github.io/.

Abstract KATAN ciphers are block ciphers using non‐linear feedback shift registers. In this study, the authors improve the results of conditional differential analysis on KATAN by using deep learning. Multi‐differential neural distinguishers are built to improve the accuracy of the neural distinguishers and increase the number of its rounds. Moreover, a conditional differential analysis framework is proposed based on deep learning with the multi‐differential neural distinguishers, resulting in a significant improvement than the previous. We present a practical key recovery attack on the 97‐round KATAN32 with 2 15.5 data complexity and 2 20.5 time complexity. The attack of the 82‐round KATAN48 and 70‐round KATAN64 are also presented as the best known practical results.

The diffusion of AI tools capable of generating realistic DeepFakes (DF) videos raises serious threats to face-based biometric recognition systems. For this reason, several detectors based on Deep Neural Networks (DNNs) have been developed to distinguish between real and DF videos. Despite their good performance, these methods suffer from vulnerability to adversarial attacks. In this paper, we argue that it is possible to increase the resilience of DNN-based DF detectors against black-box adversarial attacks by exploiting the temporal information contained in the video. By using such information, in fact, the transferability of adversarial examples from a source to a target model is significantly decreased, making it difficult to launch an attack without accessing the target network. To back this claim, we trained two convolutional neural networks (CNNs) to detect DF videos, and measured their robustness against black-box, transfer-based, attacks. We also trained two detectors by adding to the CNNs a long short-term memory (LSTM) layer to extract temporal information. Then, we measured the transferability of adversarial examples to-wards the LSTM-networks. The results we got suggest that the methods based on temporal information are less prone to black-box attacks.

Chinese hamster ovary (CHO) cell lines are widely used in industry for biological drug production. During cell culture development, considerable effort is invested to understand the factors that greatly impact cell growth, specific productivity and product qualities of the biotherapeutics. While high-throughput omics approaches have been increasingly utilized to reveal cellular mechanisms associated with cell line phenotypes and guide process optimization, comprehensive omics data analysis and management have been a challenge. Here we developed CHOmics, a web-based tool for integrative analysis of CHO cell line omics data that provides an interactive visualization of omics analysis outputs and efficient data management. CHOmics has a built-in comprehensive pipeline for RNA sequencing data processing and multi-layer statistical modules to explore relevant genes or pathways. Moreover, advanced functionalities were provided to enable users to customize their analysis and visualize the output systematically and interactively. The tool was also designed with the flexibility to accommodate other types of omics data and thereby enabling multi-omics comparison and visualization at both gene and pathway levels. Collectively, CHOmics is an integrative platform for data analysis, visualization and management with expectations to promote the broader use of omics in CHO cell research.

Recent advances in imaging genetics produce large amounts of data including functional MRI images, single nucleotide polymorphisms (SNPs), and cognitive assessments. Understanding the complex interactions among these heterogeneous and complementary data has the potential to help with diagnosis and prevention of mental disorders. However, limited efforts have been made due to the high dimensionality, group structure, and mixed type of these data. In this paper we present a novel method to detect conditional associations between imaging genetics data. We use projected distance correlation to build a conditional dependency graph among high-dimensional mixed data, then use multiple testing to detect significant group level associations (e.g., ROI-gene). In addition, we introduce a scalable algorithm based on orthogonal greedy algorithm, yielding the greedy projected distance correlation (G-PDC). This can reduce the computational cost, which is critical for analyzing large-volume of imaging genomics data. The results from our simulations demonstrate a higher degree of accuracy with GPDC than distance correlation, Pearson's correlation and partial correlation, especially when the correlation is nonlinear. Finally, we apply our method to the Philadelphia Neurodevelopmental data cohort with 866 samples including fMRI images and SNP profiles. The results uncover several statistically significant and biologically interesting interactions, which are further validated with many existing studies. The Matlab code is available at https://sites.google.com/site/jianfang86/gPDC.

BACKGROUND: With the advancement of high-throughput technologies and enrichment of popular public databases, more and more research focuses of bioinformatics research have been on computational integration of network and gene expression profiles for extracting context-dependent active subnetworks. Many methods for subnetwork searching have been developed. Scoring and searching algorithms present a range of computational considerations and implementations. The primary goal of present study is to comprehensively evaluate the performance of different subnetwork detection methods. Eleven popular methods were selected for comprehensive comparison. RESULTS: First, taking into account the dependence of genes given a protein-protein interaction (PPI) network, we simulated microarray gene expression data under case and control conditions. Then each method was applied to the simulated data for subnetwork identification. Second, a large microarray data set of prostate cancer was used to assess the practical performance of each method. Using both simulation studies and a real data application, we evaluated the performance of different methods in terms of recall and precision. CONCLUSIONS: jActiveModules, PinnacleZ and WMAXC performed well in identifying subnetwork with relative high precision and recall. BioNet performed very well only in precision. As none of methods outperformed other methods overall, users should choose an appropriate method based on the purposes of their studies.

In order to increase the accuracy for the diagnosis of schizophrenia (SCZ) disease, it is essential to integratively employ complementary information from multiple types of data. It is well known that a network is a graph based method for analyzing relationships between patients, with its nodes and edges representing patients and their relationships respectively. In this study, we developed a network-based prediction approach by taking advantage of fused network from multiple data types rather than individual networks. Specifically, we constructed a fused network using three types of data including genetic, epigenetic and neuroimaging data from the study of schizophrenia. The majority neighborhood of a node in the network was exploited for discriminating SCZ from healthy controls. In comparison with other 9 graph-based label prediction methods, our prediction method shows the best performance according to several metrics. The prediction power of our proposed method was also tested with different parameters and optimal parameters were determined. We show that the label prediction method based on network fusion from multiple data types shows promises for more accurate diagnosis of schizophrenia, which can also be extended to other disease models.

Schizophrenia (SCZ) disease ranks among the top 10 causes of disability in developed countries worldwide. Its onset is the combination result of genetic, biological and environmental factors. It is increasingly important but difficult to determine which genes are potential biomarkers for SCZ, owing to the complex nature of the pathophysiology of this disease. In our study, we integrated genomic, epigenomic and neuroimaging data to identify genetic biomarkers for schizophrenia. Important cross-correlated features were selected using multiple sparse canonical correlation analysis (smCCA) among single nucleotide polymorphism (SNP), DNA methylation and functional magnetic resonance imaging (fMRI) data. The features were then used to construct two state (health and case) gene-gene interaction networks for SNP or DNA methylation data. A network-based framework was proposed by comparing two different minimum spanning trees (MSTs), which were extracted from two fused state gene networks, respectively. We selected top 20 genes with significant changes of network features for schizophrenia. These genes were finally validated by disease association enrichment analysis, Gene Ontology (GO) enrichment analysis, pathway enrichment analysis and related literature reports. We also demonstrated the effectiveness of our framework through the comparison with other network-based discovery methods. Therefore, our proposed network-based approach can effectively discover biomarkers and resulting genes, promising better diagnosis and treatment of schizophrenia disease.

MOTIVATION: Imaging genetics combines brain imaging and genetic information to identify the relationships between genetic variants and brain activities. When the data samples belong to different classes (e.g. disease status), the relationships may exhibit class-specific patterns that can be used to facilitate the understanding of a disease. Conventional approaches often perform separate analysis on each class and report the differences, but ignore important shared patterns. RESULTS: In this paper, we develop a multivariate method to analyze the differential dependency across multiple classes. We propose a joint sparse canonical correlation analysis method, which uses a generalized fused lasso penalty to jointly estimate multiple pairs of canonical vectors with both shared and class-specific patterns. Using a data fusion approach, the method is able to detect differentially correlated modules effectively and efficiently. The results from simulation studies demonstrate its higher accuracy in discovering both common and differential canonical correlations compared to conventional sparse CCA. Using a schizophrenia dataset with 92 cases and 116 controls including a single nucleotide polymorphism (SNP) array and functional magnetic resonance imaging data, the proposed method reveals a set of distinct SNP-voxel interaction modules for the schizophrenia patients, which are verified to be both statistically and biologically significant. AVAILABILITY AND IMPLEMENTATION: The Matlab code is available at https://sites.google.com/site/jianfang86/JSCCA CONTACT: [email protected] information: Supplementary data are available at Bioinformatics online.

BACKGROUND: Integrative analysis of multi-omics data is becoming increasingly important to unravel functional mechanisms of complex diseases. However, the currently available multi-omics datasets inevitably suffer from missing values due to technical limitations and various constrains in experiments. These missing values severely hinder integrative analysis of multi-omics data. Current imputation methods mainly focus on using single omics data while ignoring biological interconnections and information imbedded in multi-omics data sets. RESULTS: In this study, a novel multi-omics imputation method was proposed to integrate multiple correlated omics datasets for improving the imputation accuracy. Our method was designed to: 1) combine the estimates of missing value from individual omics data itself as well as from other omics, and 2) simultaneously impute multiple missing omics datasets by an iterative algorithm. We compared our method with five imputation methods using single omics data at different noise levels, sample sizes and data missing rates. The results demonstrated the advantage and efficiency of our method, consistently in terms of the imputation error and the recovery of mRNA-miRNA network structure. CONCLUSIONS: We concluded that our proposed imputation method can utilize more biological information to minimize the imputation error and thus can improve the performance of downstream analysis such as genetic regulatory network construction.

Multicolor Fluorescence In-Situ Hybridization (M-FISH) is an imaging technique for rapid detection of chromosomal abnormalities, where the segmentation of chromosomes has been a challenge. Multi-channel information of M-FISH images can be used in a segmentation algorithm to exploit the correlated information across channels for better image segmentation. In addition, the neighboring pixels share similar characteristics, so this spatial information can be further utilized to improve the robustness of the algorithm to the noise. Motivated by this fact, in this paper we proposed an improved Fuzzy C-means (FCM) clustering algorithm to overcome the problems of conventional FCM such as the sensitivity to noise by incorporating both spatial and spectral information. The experimental results on both simulated and real M-FISH images have shown that our proposed method can result in higher segmentation accuracy and lower false ratio than both conventional FCM and the improved adaptive FCM (IAFCM) we recently proposed.

Schizophrenia is one of the most prevalent mental diseases, and is considered to be caused by the interplay of a number of genetic factors. In this paper, by constructing a multilayer restricted Boltzmann machines (RBMs) deep network, we use the genomic data (i.e., SNP data) for unsupervised feature learning and disease diagnosis of schizophrenia. In order to obtain some more accurate diagnosis results by RBMs, firstly, we transform the SNP data into binary sequences, and then by training the multi-layer RBMs deep network on unlabeled data, the multi-level abstract features of the genomic data are obtained and stored in the network. Finally, by adding a linear classifier to the top of the multi-layer RBMs deep network, the classification results on the testing data are gained. The results show that the average performance of this method is better than that of other methods, e.g., SVM (including linear SVM as well as SVM with multilayer perceptron kernel), sparse representations based classifier and k-nearest neighbors method. It is indicated that the multi-layer RBMs deep network can extract deep hierarchical representations of the genomic data, and then promises a more comprehensive approach for the mental disease diagnosis.

We developed a new sparse multivariate regression method, collaborative sparse reduced rank regression(C-sRRR) for detecting genetic networks associated with brain functional networks in schizophrenia (SZ). Our study: 1) introduced both genetic and brain network structure to group single nucleotide polymorphism (SNP) and voxels simultaneously for utilizing the interacting effects implied in both features; 2) used collaborative sparse group lasso to perform genetic variants selection and nuclear norm penalty to address the interrelationship among voxels; 3) developed an efficient algorithm for solving the non-smooth optimization. In real data analysis, we constructed 8605 genetic sub-networks (modules) from 722177 SNPs with a median module size of 9. A functional brain network was extracted which also showed significant discriminative characteristics between SZ and healthy controls. A sub sampling strategy was applied to identify 57 highly ranked genes from 14 high-ranking modules. 14 of them are SZ susceptibility genes and 6 genes were consistent with the findings in previous study.

BACKGROUND: The emergence of high-throughput genomic datasets from different sources and platforms (e.g., gene expression, single nucleotide polymorphisms (SNP), and copy number variation (CNV)) has greatly enhanced our understandings of the interplay of these genomic factors as well as their influences on the complex diseases. It is challenging to explore the relationship between these different types of genomic data sets. In this paper, we focus on a multivariate statistical method, canonical correlation analysis (CCA) method for this problem. Conventional CCA method does not work effectively if the number of data samples is significantly less than that of biomarkers, which is a typical case for genomic data (e.g., SNPs). Sparse CCA (sCCA) methods were introduced to overcome such difficulty, mostly using penalizations with l-1 norm (CCA-l1) or the combination of l-1and l-2 norm (CCA-elastic net). However, they overlook the structural or group effect within genomic data in the analysis, which often exist and are important (e.g., SNPs spanning a gene interact and work together as a group). RESULTS: We propose a new group sparse CCA method (CCA-sparse group) along with an effective numerical algorithm to study the mutual relationship between two different types of genomic data (i.e., SNP and gene expression). We then extend the model to a more general formulation that can include the existing sCCA models. We apply the model to feature/variable selection from two data sets and compare our group sparse CCA method with existing sCCA methods on both simulation and two real datasets (human gliomas data and NCI60 data). We use a graphical representation of the samples with a pair of canonical variates to demonstrate the discriminating characteristic of the selected features. Pathway analysis is further performed for biological interpretation of those features. CONCLUSIONS: The CCA-sparse group method incorporates group effects of features into the correlation analysis while performs individual feature selection simultaneously. It outperforms the two sCCA methods (CCA-l1 and CCA-group) by identifying the correlated features with more true positives while controlling total discordance at a lower level on the simulated data, even if the group effect does not exist or there are irrelevant features grouped with true correlated features. Compared with our proposed CCA-group sparse models, CCA-l1 tends to select less true correlated features while CCA-group inclines to select more redundant features.

It is important to identify significant biomarkers such as SNPs for medical diagnosis and treatment. However, the size of a biological sample is usually far less than the number of measurements, which makes the problem more challenging. To overcome this difficulty, we propose a sparse representation based variable selection (SRVS) approach. A simulated data set was first tested to demonstrate the advantages and properties of the proposed method. Then, we applied the algorithm to a joint analysis of 759075 SNPs and 153594 functional magnetic resonance imaging (fMRJ) voxels in 208 subjects (92 cases/116 controls) to identify significant biomarkers for schizophrenia (SZ). When compared with previous studies, our proposed method located 20 genes out of the top 45 SZ genes that are publicly reported We also detected some interesting functional brain regions from the fMRI study. In addition, a leave one out (LOO) cross-validation was performed and the results were compared with that of a previously reported method, which showed that our method gave significantly higher classification accuracy. In addition, the identification accuracy with integrative analysis is much better than that of using single type of data, suggesting that integrative analysis may lead to better diagnostic accuracy by combining complementary SNP and fMRI data.

We developed a structure based sparse representation model for classifying chromosomes in M-FISH images. The sparse representation based classification model used in our previous work only considered one pixel without incorporating any structural information. The new proposed model extends the previous one to multiple pixels case, where each target pixel together with its neighboring pixels will be used simultaneously for classification. We also extend Orthogonal Matching Pursuit (OMP) algorithm to the multiple pixels case, named simultaneous OMP algorithm (SOMP), to solve the structure based sparse representation model. The classification results show that our new model outperforms the previous sparse representation model with the p-value less than le-6. We also discussed the effects of several parameters (neighborhood size, sparsity level, and training sample size) on the accuracy of the classification. Our proposed method can be affected by the sparsity level and the neighborhood size but is insensitive to the training sample size. Therefore, the comparison indicates that the structure based sparse representation model can significantly improve the accuracy of the chromosome classification, leading to improved diagnosis of genetic diseases and cancers.

We designed a sparse representation clustering (SRC) model to select the significant single nucleotide polymorphisms (SNPs) and proposed a novel SRC with a sliding window model for functional magnetic resonance imaging (fMRI) voxels selection. Then we combined two types of data to classify schizophrenia patients from healthy controls by linear support vector machine (SVM) to achieve a better diagnosis of schizophrenia. The effectiveness of the selected variables (SNPs or voxels) was validated by the leave one out (LOO) cross-validation method. The experimental results show that our proposed SRC method can effectively select the most discriminative variables in both SNPs and fMRI data. In particular, the combination of complementary fMRI and SNP data can significantly improve the classification of schizophrenia patients, which provides new insights in the study of schizophrenia.