Cancer is a complex disease characterized by uncontrolled cell growth and requires an accurate classification for effective treatment. T cell receptors (TCRs), crucial proteins in the immune system, play a pivotal role in antigen recognition. Advancements in sequencing technologies have facilitated the comprehensive profiling of TCR repertoires, uncovering TCRs with potent anticancer activity and enabling TCR-based immunotherapies. Performing an effective analysis of these complex biomolecules requires representations that accurately capture both their structural and functional characteristics. T cell protein sequences pose unique challenges because of their relatively shorter lengths compared to other biomolecules. Traditional vector-based embedding methods may encounter issues such as information loss. Therefore, an image-based representation approach becomes a preferred choice for efficient embedding, allowing the preservation of essential details and enabling a comprehensive analysis of T cell protein sequences. We propose generating images from protein sequences using the concept of chaos game representation (CGR). We design images using the kaleidoscopic images approach. This Deep Learning-Assisted Analysis of ProteiN Sequences Using Chaos Enhanced Kaleidoscopic Images (called DANCE) provides a unique way to visualize protein sequences by recursively applying chaos game rules around a central seed point. The resulting kaleidoscopic images exhibit symmetrical patterns that offer a visual representation of the protein sequences. To investigate the effectiveness of this approach, we perform classification of the TCR protein sequences in terms of their respective target cancer cells, since TCRs are known for their immune response against cancer disease. The DANCE technique is used to turn the TCR sequences into pictures before classification. We employ deep learning (DL) vision models to classify the generated images to obtain insight into the relationship between the visual patterns in the generated kaleidoscopic images and the underlying protein properties. By combining CGR-based image generation with DL classification, this study opens new possibilities in protein analysis.

Anti-cancer peptide (ACP) sequence classification is crucial for cancer treatment development. Current neural network approaches achieve high accuracy but require substantial parameters and training data. Recent compression-based methods compress entire sequences, potentially missing fine grained neighboring information critical for classification. We propose a novel approach integrating Gzip compression with an incremental k-mer strategy. Unlike conventional methods, we compress individual k-mers and incrementally build subsequence compressions, preserving amino acid-level context. Using Normalized Compression Distance (NCD) and kernel-based embeddings, our parameter-free method achieves state-of-the-art performance on breast and lung cancer ACP datasets, outperforming deep neural networks and large language models without requiring custom features or pre-trained models. Our approach provides a practical, efficient alternative to computationally intensive methods, proving effective even in low-resource environments.

Equivariant graph neural network (GNN) methods for antibody complementarity-determining region (CDR) design achieve the highest sequence recovery but suffer from severe vocabulary collapse. The current best GNN methods over-predict very few amino acids, such as tyrosine and glycine, while ignoring functionally important residues. We trace this failure to GNN encoders learning amino acid distributions de novo from limited structural data, discarding substitution patterns encoded in evolutionary databases. To resolve this, we propose EvoStruct, which bridges a frozen protein language model (PLM) with 3D structural context from an E(3)-equivariant GNN via a cross-attention adapter. Unlike prior PLM-structure adapters for general protein design, EvoStruct targets the vocabulary collapse problem specific to CDR design through progressive PLM unfreezing and R-Drop consistency regularization. On the CHIMERA-Bench dataset, EvoStruct achieves the highest amino acid recovery and lowest perplexity among several antibody design methods, improving sequence recovery by 16

Antibody design methods condition on antigen structure to generate complementarity-determining regions (CDR), yet a systematic evaluation of baseline methods reveals that they largely ignore the antigen input. We identify three failure modes that explain this behavior. Antigen blindness arises because models derive predictions from antibody framework context rather than antigen information, producing nearly identical CDRs regardless of the target. Vocabulary collapse reduces predicted amino acids to three to five per position, far below the ground truth distribution in native sequences. Moreover, any model trained with standard per-position cross-entropy converges to the positional marginal distribution, making it provably unable to produce antigen-specific sequence predictions. We propose a novel encoder-decoder architecture called AgForce, that uses a graph neural network (GNN) as the encoder and specialized decoders for sequence-structure co-design. Specifically, we apply framework dropout, gated bottlenecks, and hyperbolic cross attention that prevent the antibody shortcut path. In the decoder, a Mixture Density Network (MDN) sequence head with Potts-like pairwise coupling and annealed Multiple Choice Learning (aMCL) replaces the cross-entropy objective with a multi-component distribution whose optimal solution differs from the positional marginal. An antigen cycle consistency head routes gradients through the sequence decoder, forcing predicted distributions to encode antigen identity. AgForce achieves the best binding quality and sequence recovery simultaneously on the CHIMERA-Bench dataset, improving amino acid recovery by 8

This work concerns the harmonization of geospatial data to improve linkages between place-based characteristics and health outcomes. Such data are typically available as geographic layers, each representing a distinct attribute (e.g., income or distance to a clinic). Since layers are typically constructed independently, their boundaries tend to be spatially incongruent, which can create inconsistencies and introduce bias. This motivates developing algorithmic approaches for aligning such layers while aiming to preserve spatial integrity. This paper formalizes the problem of aligning k collections of m spatial supports over n spatial units in a d-dimensional Euclidean space such that maximum distortion to any collection is minimized. In the above setting, k is the number of layers; n is an indivisible population unit (e.g., census tract); m denotes supports, which are larger regions aggregating a set of contiguous units in order to capture broader regional patterns or enhance statistical stability; and d=2. It is shown that: (1) the one-dimensional case is solvable in time polynomial in k, m, and n; (2) the two-dimensional case is NP-hard for two collections of two supports each; and (3) a heuristic can be provided for aligning a set of collections in the two-dimensional case, which is of practical importance.

Computational antibody CDR design methods condition on antigen structure to generate binding loops, yet existing architectures conflate two fundamentally distinct sub-problems: identifying which CDR positions will contact the antigen, and selecting amino acids at those positions. This conflation forces models to learn contact reasoning implicitly through uniform message passing, diluting antigen signal across all positions equally. We introduce ConTact, a contact-then-act architecture that explicitly decomposes CDR design into three cascaded stages: learning surface complementarity fingerprints, predicting CDR-antigen contacts, and injecting contact-gated antigen features into the sequence head. A distance-biased cross-attention module encodes geometric priors favoring spatial neighbors, while a contact-weighted cross-entropy loss concentrates gradient signal on binding-critical positions. On CHIMERA-Bench dataset, ConTact achieves the best structural quality (7

Understanding epistatic interactions, where mutations collectively influence viral fitness, is critical for predicting pathogen evolution. We present a hidden Markov model (HMM) framework that captures the temporal dynamics of epistatic relationships in SARS-CoV-2, addressing limitations of static network-based approaches. Our method models single amino acid variant pairs as a two-state system (linked/unlinked), with emission probabilities derived from linkage disequilibrium theory and transition probabilities optimized via the Baum-Welch algorithm. We implement permutation-based validation with temporal order noise reduction (>80% agreement across five iterations) to distinguish biological signals from stochastic noise. Applied to 2,192,008 spike protein sequences from the United States (March 2020-December 2021), our approach identified three classes of epistatic dynamics: permanent (0.3%), transient (0.3%), and oscillating (0.7%) linkages. Analysis of Alpha variant positions revealed 78% epistatic linkage compared with 1.3% across all spike protein position pairs, with 60% exhibiting oscillating patterns suggestive of frequency-dependent selection. We detected all 17 previously reported epistatic pairs plus 18 novel interactions, including critical connections between positions 69-70 and other functional sites. Notably, Alpha variant epistatic networks were detectable as early as April 2020, months before widespread circulation. Our framework scales to variant-wide analysis, revealing distinct patterns across variants: Delta (96% linkage, 71% oscillating) and Omicron (87% linkage, 56% oscillating). The computational pipeline, implemented with parallelized HMM training and Viterbi decoding, processes hundreds of thousands of position pairs efficiently. By transforming epistasis detection from static to temporal analysis, this work provides computational tools for early variant detection and demonstrates how probabilistic modeling can capture evolutionary dynamics in real-time genomic surveillance systems.

Antibodies neutralize foreign antigens by binding to specific surface regions called epitopes. Computational epitope prediction is critical for understanding immune recognition and guiding antibody engineering. However, existing methods face three fundamental challenges: antibody-aware models encode each chain independently and combine them only at a late stage, failing to capture co-dependent structural features that define binding interfaces, whereas severe class imbalance and scarcity of known antibody-antigen complexes render standard training objectives ineffective. We propose EpiFormer, a general encoder-decoder framework that addresses these challenges jointly. Our key design principle is interleaved cross-attention within GNN encoding layers, enabling bidirectional antigen-antibody information flow throughout representation learning rather than only at the output. This early-fusion principle is backbone-agnostic, providing consistent gains across GNN architectures from simple GCNs to equivariant models. We further show that sparsity-aware objectives are effective when paired with early-fusion architectures for the epitope prediction task. EpiFormer improves over the previous best method by over 40

Discrete diffusion and flow-matching models denoise a sequence over many steps, but to keep each step cheap, they factorize the transition across positions and decide every token independently. This makes few-step generation challenging for text when the target couples two positions, such as a subject and a verb that must agree. An independent update commits to them separately, and many function evaluations are spent repairing the mismatch. Existing few-step methods buy back the lost correlation by distilling or rectifying a slow teacher, and so inherit the teacher's quality ceiling. We ask instead whether a model can express correlated steps natively, and answer with Latent-Kernel Discrete Flow Maps (LKF), a from-scratch flow-map kernel that is a mixture of M factorized components tied by a single shared latent. Conditioned on the latent, each component is cheap, and the mixture is summed over the latent in closed form for small M. We show that a single step places mass on correlated completions with the same sampling time complexity as a factorized model, since one latent is drawn per sequence and reused across the entire denoising trajectory. We also show that the Masked Diffusion Language Model (MDLM) is a special case of our LKF model at M=1. The experiments for unconditional text generation on the One-Billion-Word (LM1B) and WikiText-103 benchmarks show that our LKF model learns strongly heterogeneous components and improves generative perplexity by 2.1x to 3.3x over the likelihood baselines without losing diversity. The gain grows with M, and at M=8, it surpasses distilled and rectified few-step samplers. The source code is available at: https://github.com/mansoor181/lkf.git

Molecular sequence analysis is crucial for understanding several biological processes, including protein-protein interactions, functional annotation, and disease classification. The large number of sequences and the inherently complicated nature of protein structures make it challenging to analyze such data. Finding patterns and enhancing subsequent research requires the use of dimensionality reduction and feature selection approaches. Recently, a method called Correlated Clustering and Projection (CCP) has been proposed as an effective method for biological sequencing data. The CCP technique remains computationally expensive, despite its effectiveness for sequence visualization. Furthermore, its utility for classifying molecular sequences is still uncertain. To solve these two problems, we present a Nearest-Neighbor Correlated Clustering and Projection (CCP-NN)-based technique for efficiently preprocessing molecular sequence data. To group related molecular sequences and produce representative supersequences, CCP makes use of sequence-to-sequence correlations. As opposed to conventional methods, CCP does not rely on matrix diagonalization, therefore, it can be applied to a range of machine-learning problems. We estimate the density map and compute the correlation using a nearest-neighbor search technique. We perform a molecular sequence classification using CCP and CCP-NN representations to assess the efficacy of our proposed approach. Our findings show that CCP-NN considerably improves classification accuracy and significantly outperforms CCP in computational runtime.

Computational antibody design has seen rapid methodological progress, with dozens of deep generative methods proposed in the past three years, yet the field lacks a standardized benchmark for fair comparison and model development. These methods are evaluated on different SAbDab snapshots, non-overlapping test sets, and incompatible metrics, and the literature fragments the design problem into numerous sub-tasks with no common definition. We introduce Chimera-Bench (CDR Modeling with Epitope-guided Redesign), a unified benchmark built around a single canonical task: epitope-conditioned CDR sequence-structure co-design. Chimera-Bench provides (1) a curated, deduplicated dataset of 2,922 antibody-antigen complexes with epitope and paratope annotations; (2) three biologically motivated splits testing generalization to unseen epitopes, unseen antigen folds, and prospective temporal targets; and (3) a comprehensive evaluation protocol with five metric groups including novel epitope-specificity measures. We benchmark representative methods spanning different generative paradigms and report results across all splits. Chimera-Bench is the largest dataset of its kind for the antibody design problem, allowing the community to develop and test novel methods and evaluate their generalizability. The source code and data are available at: https://github.com/mansoor181/chimera-bench.git

The rapid advancement of large language models (LLMs) has enabled significant strides in various fields. This paper introduces a novel approach to evaluate the effectiveness of LLM embeddings in the context of inherent geometric properties. We investigate the structural properties of these embeddings through three complementary metrics δ-hyperbolicity, Ultrametricity, and Neighbor Joining. δ-hyperbolicity, a measure derived from geometric group theory, quantifies how much a metric space deviates from being a tree-like structure. In contrast, ultrametricity characterizes strictly hierarchical structures where distances obey a strong triangle inequality. While Neighbor Joining quantifies how tree-like the distance relationships are, it does so specifically with respect to the tree reconstructed by the Neighbor Joining algorithm. By analyzing the embeddings generated by LLMs using these metrics, we uncover to what extent the embedding space reflects an underlying hierarchical or tree-like organization. Our findings reveal that LLM embeddings exhibit varying degrees of hyperbolicity and ultrametricity, which correlate with their performance in the underlying machine learning tasks.

The application of Digital Twin (DT) technology and Federated Learning (FL) has great potential to change the field of biomedical image analysis, particularly for Computed Tomography (CT) scans. This paper presents Federated Transfer Learning (FTL) as a new Digital Twin-based CT scan analysis paradigm. FTL uses pre-trained models and knowledge transfer between peer nodes to solve problems such as data privacy, limited computing resources, and data heterogeneity. The proposed framework allows real-time collaboration between cloud servers and Digital Twin-enabled CT scanners while protecting patient identity. We apply the FTL method to a heterogeneous CT scan dataset and assess model performance using convergence time, model accuracy, precision, recall, F1 score, and confusion matrix. It has been shown to perform better than conventional FL and Clustered Federated Learning (CFL) methods with better precision, accuracy, recall, and F1-score. The technique is beneficial in settings where the data is not independently and identically distributed (non-IID), and it offers reliable, efficient, and secure solutions for medical diagnosis. These findings highlight the possibility of using FTL to improve decision-making in digital twin-based CT scan analysis, secure and efficient medical image analysis, promote privacy, and open new possibilities for applying precision medicine and smart healthcare systems.

SMILES (Simplified Molecular Input Line Entry System) strings are widely used to represent molecular structures in cheminformatics and drug discovery. However, effectively transforming these string-based representations into meaningful numerical features for machine learning remains a significant challenge due to the complex, non-Euclidean nature of molecular structures. Traditional fingerprint-based and deep learning approaches often struggle with scalability, interpretability, or computational efficiency. Our approach leverages the Morgan Fingerprint to generate molecular feature representations, followed by a pairwise kernel function to compute a structured similarity matrix. We then refine this matrix using the Sinkhorn-Knopp algorithm, ensuring it satisfies probabilistic constraints. To reduce dimensionality, we apply Kernel Principal Component Analysis (PCA), producing compact embeddings suitable for downstream machine learning tasks. We conduct a comprehensive empirical evaluation of the proposed method which is assessed for drug subcategory prediction (classification task) and solubility AlogPS “aqueous solubility and octanol/water partition coefficient” (regression task) using the benchmark SMILES string dataset. The outcomes show the proposed method outperforms baseline methods in supervised analysis and has potential uses in molecular design and drug discovery. By integrating kernel-based learning with probabilistic refinement, our method offers a promising alternative to existing cheminformatics techniques.

Understanding the structural and functional characteristics of proteins are crucial for developing preventative and curative strategies that impact fields from drug discovery to policy development. An important and popular technique for examining how amino acids make up these characteristics of the protein sequences with position-specific scoring (PSS). While the string kernel is crucial in natural language processing (NLP), it is unclear if string kernels can extract biologically meaningful information from protein sequences, despite the fact that they have been shown to be effective in the general sequence analysis tasks. In this work, we propose a weighted PSS kernel matrix (or W-PSSKM), that combines a PSS representation of protein sequences, which encodes the frequency information of each amino acid in a sequence, with the notion of the string kernel. This results in a novel kernel function that outperforms many other approaches for protein sequence classification. We perform extensive experimentation to evaluate the proposed method. Our findings demonstrate that the W-PSSKM significantly outperforms existing baselines and state-of-the-art methods in terms of predictive performance.

Cardiac arrest is one of the biggest global health problems, and early identification and management are key to enhancing the patient's prognosis. In this paper, we propose a novel framework that combines an EfficientNet-based deep learning model with a digital twin system to improve the early detection and analysis of cardiac arrest. We use compound scaling and EfficientNet to learn the features of cardiovascular images. In parallel, the digital twin creates a realistic and individualized cardiovascular system model of the patient based on data received from the Internet of Things (IoT) devices attached to the patient, which can help in the constant assessment of the patient and the impact of possible treatment plans. As shown by our experiments, the proposed system is highly accurate in its prediction abilities and, at the same time, efficient. Combining highly advanced techniques such as deep learning and digital twin (DT) technology presents the possibility of using an active and individual approach to predicting cardiac disease.

The availability of SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) virus data post-COVID has reached exponentially to an enormous magnitude, opening research doors to analyze its behavior. Various studies are conducted by researchers to gain a deeper understanding of the virus, like genomic surveillance, etc, so that efficient prevention mechanisms can be developed. However, the unstable nature of the virus (rapid mutations, multiple hosts, etc) creates challenges in designing analytical systems for it. Therefore, we propose a neural network-based (NN) mechanism to perform an efficient analysis of the SARS-CoV-2 data, as NN portrays generalized behavior upon training. Moreover, rather than using the full-length genome of the virus, we apply our method to its spike region, as this region is known to have predominant mutations and is used to attach to the host cell membrane. In this paper, we introduce a pipeline that first converts the spike protein sequences into a fixed-length numerical representation and then uses Neuromorphic Spiking Neural Network to classify those sequences. We compare the performance of our method with various baselines using real-world SARS-CoV-2 spike sequence data and show that our method is able to achieve higher predictive accuracy compared to the recent baselines.

A crucial challenge in molecular biology is the prediction of DNA-protein binding interactions, which has applications in the study of gene regulation and genome functionality. In this paper, we present a novel approach involving hashing-based embedding to predict DNA-protein binding interactions with increased accuracy and faster computation speed. Before applying machine learning models for predicting DNA-protein binding, we need to obtain an appropriate numeric representation of the biological sequences. There exist several embedding options for protein sequences but the computation cost allocated with them is high. Therefore, in our approach, we propose hashing-based embedding, which is alignment-free, has faster generation speed, and is computationally inexpensive as compared to baseline embedding methods. Our approach captures the sequence-specific binding preferences of proteins, enabling researchers to learn more about the underlying binding mechanisms. Experiments on diverse DNA-protein interaction datasets demonstrate that the proposed approach not only improves the predictive performance in terms of accuracy but is also much more efficient as compared to the baseline methods. The proposed method holds significant potential in deciphering intricate DNA-protein interactions, ultimately advancing our comprehension of gene regulation mechanisms.