How a single fertilized cell gives rise to a complex array of specialized cell types in development is a central question in biology. The cells replicate to generate cell lineages and acquire differentiated characteristics through poorly understood molecular processes. A key approach to studying developmental processes is to infer the tree graph of cell lineage histories, which provides an analytical framework for dissecting individual cells' molecular decisions during replication and differentiation (i.e., acquisition of specialized traits). Although genetically engineered lineage-tracing methods have advanced the field, they are either infeasible or ethically constrained in many organisms. By contrast, modern single-cell technologies can measure high-content molecular profiles (*e.g.*, transcriptomes) in a wide range of biological systems. Here, we introduce *CellTreeQM*, a novel deep learning method based on transformer architectures that learns an embedding space with geometric properties optimized for tree-graph inference. By formulating the lineage reconstruction problem as tree-metric learning, we systematically explore weakly supervised training settings at different levels of information and present the *Cell Lineage Reconstruction Benchmark* to facilitate comprehensive evaluation. This benchmark includes (1) synthetic data modeled via Brownian motion with independent noise and spurious signals; (2) lineage-resolved single-cell RNA sequencing datasets. Experimental results show that *CellTreeQM* recovers lineage structures with minimal supervision and limited data, offering a scalable framework for uncovering cell lineage relationships. To our knowledge, this is the first method to cast cell lineage inference explicitly as a metric learning task, paving the way for future computational models aimed at uncovering the molecular dynamics of cell lineage. Code and benchmarks are available at: https://kuang-da.github.io/CellTreeQM-page

As a new paradigm of distributed learning, Federated Learning (FL) has been applied in industrial fields, such as intelligent retail, finance and autonomous driving. However, several schemes that aim to attack robust aggregation rules and reducing the model accuracy have been proposed recently. These schemes do not maintain the sign statistics of gradients unchanged during attacks. Therefore, the sign statistics-based scheme SignGuard can resist most existing attacks. To defeat SignGuard and most existing cosine or distance-based aggregation schemes, we propose an enhanced model poisoning attack, ScaleSign. Specifically, ScaleSign uses a scaling attack and a sign modification component to obtain malicious gradients with higher cosine similarity and modify the sign statistics of malicious gradients, respectively. In addition, these two components have the least impact on the magnitudes of gradients. Then, we propose MSGuard, a Multi-Strategy Byzantine-robust scheme based on cosine mechanisms, symbol statistics, and spectral methods. Formal analysis proves that malicious gradients generated by ScaleSign have a closer cosine similarity than honest gradients. Extensive experiments demonstrate that ScaleSign can attack most of the existing Byzantine-robust rules, especially achieving a success rate of up to 98.23% for attacks on SignGuard. MSGuard can defend against most existing attacks including ScaleSign. Specifically, in the face of ScaleSign attack, the accuracy of MSGuard improves by up to 41.78% compared to SignGuard.

Existing research often posits spurious features as **easier** to learn than core features in neural network optimization, but the impact of their relative simplicity remains under-explored. Moreover, studies mainly focus on end performance rather than the learning dynamics of feature learning. In this paper, we propose a theoretical framework and an associated synthetic dataset grounded in boolean function analysis. This setup allows for fine-grained control over the relative complexity (compared to core features) and correlation strength (with respect to the label) of spurious features to study the dynamics of feature learning under spurious correlations. Our findings uncover several interesting phenomena: (1) stronger spurious correlations or simpler spurious features slow down the learning rate of the core features, (2) two distinct subnetworks are formed to learn core and spurious features separately, (3) learning phases of spurious and core features are not always separable, (4) spurious features are not forgotten even after core features are fully learned. We demonstrate that our findings justify the success of retraining the last layer to remove spurious correlation and also identifies limitations of popular debiasing algorithms that exploit early learning of spurious features. We support our empirical findings with theoretical analyses for the case of learning XOR features with a one-hidden-layer ReLU network.

To address the challenges posed by data silos and heterogeneity in distributed machine learning, privacy-preserving asynchronous Federated Learning (FL) has been extensively explored in academic and industrial fields. However, existing privacy-preserving asynchronous FL schemes still suffer from the problem of low model accuracy caused by inconsistency between delayed model updates and current model updates, and even cannot adapt well to Non-Independent and Identically Distributed (Non-IID) settings. To address these issues, we propose a Privacy-preserving Asynchronous Federated Learning based on the alternating direction multiplier method (PAFed), which is able to achieve high-accuracy models in Non-IID settings. Specifically, we utilize vector projection techniques to correct the inconsistency between delayed model updates and current model updates, thereby reducing the impact of delayed model updates on the aggregation of current model updates. Additionally, we employ an optimization method based on alternating direction multipliers to adapt the Non-IID settings to further enhance the global model accuracy. Finally, through extensive experiments, we demonstrate that our scheme improves the model accuracy by up to 12.53% when compared with current state-of-the-art solution FedADMM.

To achieve the search over encrypted data in cloud server, Searchable Encryption (SE) has attracted extensive attention from both academic and industrial fields. The existing Bloom filter-based SE schemes can achieve similarity search, but will generally incur high false positive rates, and even leak the privacy of values in Bloom filters (BF). To solve the above problems, we first propose a basicPrivacy-preservingBloom filter-basedKeywordSearch scheme using the Circular Shift and Coalesce-Bloom Filter (CSC-BF) and Symmetric-key Hidden Vector Encryption (SHVE) technology (namely PBKS), which can achieve effective search while protecting the values in BFs. Then, we design a new index structure T-CSCBF utilizing theTwin Bloom Filter (TBF) technology. Based on this, we propose an improved scheme PBKS+, which assigns a unique inclusion identifier to each position in each BF with privacy protection. Formal security analysis proves that our schemes are secure against Indistinguishability under Selective Chosen-Plaintext Attack (IND-SCPA), and extensive experiments using real-world datasets demonstrate that our schemes are feasible in practice.

MOTIVATION: When rare missense variants are clinically interpreted as to their pathogenicity, most are classified as variants of uncertain significance (VUS). Although functional assays can provide strong evidence for variant classification, such results are generally unavailable. Multiplexed assays of variant effect can generate experimental 'variant effect maps' that score nearly all possible missense variants in selected protein targets for their impact on protein function. However, these efforts have not always prioritized proteins for which variant effect maps would have the greatest impact on clinical variant interpretation. RESULTS: Here, we mined databases of clinically interpreted variants and applied three strategies, each building on the previous, to prioritize genes for systematic functional testing of missense variation. The strategies ranked genes (i) by the number of unique missense VUS that had been reported to ClinVar; (ii) by movability- and reappearance-weighted impact scores, to give extra weight to reappearing, movable VUS and (iii) by difficulty-adjusted impact scores, to account for the more resource-intensive nature of generating variant effect maps for longer genes. Our results could be used to guide systematic functional testing of missense variation toward greater impact on clinical variant interpretation. AVAILABILITY AND IMPLEMENTATION: Source code available at: https://github.com/rothlab/mave-gene-prioritization. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Multiplexed assays of variant effect (MAVEs) are capable of experimentally testing all possible single nucleotide or amino acid variants in selected genomic regions, generating 'variant effect maps', which provide biochemical insight and functional evidence to enable more rapid and accurate clinical interpretation of human variation. Because the international community applying MAVE approaches is growing rapidly, we developed the online MaveRegistry platform to catalyze collaboration, reduce redundant efforts, allow stakeholders to nominate targets and enable tracking and sharing of progress on ongoing MAVE projects. AVAILABILITY AND IMPLEMENTATION: MaveRegistry service: https://registry.varianteffect.org. MaveRegistry source code: https://github.com/kvnkuang/maveregistry-front-end.

All mRNA quantification, including single-cell mRNA sequencing, requires normalization to correct for technical variation and to make measurements of two or more samples comparable. The choice of normalization method impacts the downstream analysis. All common approaches (applying scaling factors, variational inference, and quantile regression) currently focus on removing technical variations but ignore localized variations of biological origin. To address this problem, we propose a new framework to normalize for technical effects while also aligning RNA-seq datasets for a biologically meaningful comparison. We present an iterative optimization method using the notion of a resistant fit regression to isolate localized perturbations. Both simulated data and real data are resistant-fit normalized and compared with popular normalization methods. This comparison shows that the resistant fit works better under localized biological variations.

SUMMARY: Fully realizing the promise of personalized medicine will require rapid and accurate classification of pathogenic human variation. Multiplexed assays of variant effect (MAVEs) can experimentally test nearly all possible variants in selected gene targets. Planning a MAVE study involves identifying target genes with clinical impact, and identifying scalable functional assays for that target. Here, we describe MaveQuest, a web-based resource enabling systematic variant effect mapping studies by identifying potential functional assays, disease phenotypes and clinical relevance for nearly all human protein-coding genes. AVAILABILITY AND IMPLEMENTATION: MaveQuest service: https://mavequest.varianteffect.org/. MaveQuest source code: https://github.com/kvnkuang/mavequest-front-end/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

In this paper, a graph-based nonlocal total variation method is proposed for unsupervised classification of hyperspectral images (HSI). The variational problem is solved by the primal-dual hybrid gradient algorithm. By squaring the labeling function and using a stable simplex clustering routine, an unsupervised clustering method with random initialization can be implemented. The effectiveness of this proposed algorithm is illustrated on both synthetic and real-world HSI, and numerical results show that the proposed algorithm outperforms other standard unsupervised clustering methods, such as spherical K -means, nonnegative matrix factorization, and the graph-based Merriman-Bence-Osher scheme.

In recent years, several feature encoding schemes for the bags-of-visual-words model have been proposed. While most of these schemes produce impressive results, they all share an important limitation: their high computational complexity makes it challenging to use them for large-scale problems. In this work, we propose an approximate locality-constrained encoding scheme that offers significantly better computational efficiency (~ 40×) than its exact counterpart, with comparable classification accuracy. Using the perturbation analysis of least-squares problems, we present a formal approximation error analysis of our approach, which helps distill the intuition behind the robustness of our method. We present a thorough set of empirical analyses on multiple standard data-sets, to assess the capability of our encoding scheme for its representational as well as discriminative accuracy.

In this paper, we design a fast hierarchical clustering algorithm for high-resolution hyperspectral images (HSI). At the core of the algorithm, a new rank-two nonnegative matrix factorization (NMF) algorithm is used to split the clusters, which is motivated by convex geometry concepts. The method starts with a single cluster containing all pixels and, at each step, performs the following: 1) selects a cluster in such a way that the error at the next step is minimized and 2) splits the selected cluster into two disjoint clusters using rank-two NMF in such a way that the clusters are well balanced and stable. The proposed method can also be used as an endmember extraction algorithm in the presence of pure pixels. The effectiveness of this approach is illustrated on several synthetic and real-world HSIs and is shown to outperform standard clustering techniques such as k-means, spherical k-means, and standard NMF.

Nonnegative matrix factorization (NMF) has been successfully used as a clustering method especially for flat partitioning of documents. In this paper, we propose an efficient hierarchical document clustering method based on a new algorithm for rank-2 NMF. When the two block coordinate descent framework of nonnegative least squares is applied to computing rank-2 NMF, each subproblem requires a solution for nonnegative least squares with only two columns in the matrix. We design the algorithm for rank-2 NMF by exploiting the fact that an exhaustive search for the optimal active set can be performed extremely fast when solving these NNLS problems. In addition, we design a measure based on the results of rank-2 NMF for determining which leaf node should be further split. On a number of text data sets, our proposed method produces high-quality tree structures in significantly less time compared to other methods such as hierarchical K-means, standard NMF, and latent Dirichlet allocation.

Nonnegative matrix factorization (NMF) provides a lower rank approximation of a nonnegative matrix, and has been successfully used as a clustering method. In this paper, we offer some conceptual understanding for the capabilities and shortcomings of NMF as a clustering method. Then, we propose Symmetric NMF (SymNMF) as a general framework for graph clustering, which inherits the advantages of NMF by enforcing nonnegativity on the clustering assignment matrix. Unlike NMF, however, SymNMF is based on a similarity measure between data points, and factorizes a symmetric matrix containing pairwise similarity values (not necessarily nonnegative). We compare SymNMF with the widely-used spectral clustering methods, and give an intuitive explanation of why SymNMF captures the cluster structure embedded in the graph representation more naturally. In addition, we develop a Newton-like algorithm that exploits second-order information efficiently, so as to show the feasibility of SymNMF as a practical framework for graph clustering. Our experiments on artificial graph data, text data, and image data demonstrate the substantially enhanced clustering quality of SymNMF over spectral clustering and NMF. Therefore, SymNMF is able to achieve better clustering results on both linear and nonlinear manifolds, and serves as a potential basis for many extensions and applications.

The original Yahoo! search engine consists of manually organized topic hierarchy of webpages for easy browsing. Modern search engines (such as Google and Bing), on the other hand, return a flat

Most of the multi-objective genetic algorithms (MOGAs) can be divided into two steps, namely constructing the nondominated set and truncation procedure. The quality of the latter directly affects the efficiency and the distribution of MOGA. In this paper, a new MOGA named PCGA (polar coordinates genetic algorithm) is proposed. The technique, which uses grids to keep diversity of solutions with polar coordinates, is introduced into PCGA. The time complexity of its truncation approach is higher than that of NSGA2, but is greatly lower than that of SPEA2. Meanwhile, though PCGA's distribution is not as good as that of SPEA2, it makes a large improvement with respect to that of NSGA2.