arXiv subjects

Sarwan Ali

Publications and source records attributed to Sarwan Ali.

At least 19 recordsLinked to original sources

arXiv2026

Causal dictionary learning reveals and validates transcription-factor binding features in genomic language models

Genomic language models achieve strong performance across regulatory-genomics tasks, yet what these models internally represent remains opaque, and the field lacks a principled procedure for verifying that an apparent ``concept'' inside a model is real rather than an artifact of sequence composition. We introduce a framework that combines sparse dictionary learning with causal intervention to extract, validate, and causally test interpretable features in genomic foundation models. Training top-$k$ sparse autoencoders on the hidden activations of two architecturally distinct models, Nucleotide Transformer ($6$-mer tokenization) and DNABERT-2 (byte-pair encoding), we recover thousands of monosemantic features that map to transcription-factor (TF) sequence motifs. We show that the naive validation of such features against position weight matrices is severely confounded by GC composition and repetitive elements, producing hundreds of spurious ``TF features'', and we develop a composition-matched, binding-resolved protocol that removes these confounds. Critically, we move beyond correlation: by ablating individual dictionary directions during the model's forward pass and measuring the induced shift in the model's own predictive distribution, we establish that specific features are \emph{causally} used to represent cell-type-specific TF binding, not merely motif presence. Across three transcription factors (CTCF, GATA1, REST) and both architectures, causally validated binding features emerge reproducibly ($7$--$14$ of $15$ tested features per condition), while two classes of negative control, scrambled binding labels and randomly selected features, yield no detectable signal. The framework is purely computational, uses only public data, and provides a reusable standard for interpretability claims in genomic deep learning.

q-bio.GN↗

arXiv2026

Auditing pretraining contamination in single-cell foundation model benchmarks

Single-cell foundation models (scFMs) such as Geneformer, scGPT, and Universal Cell Embeddings (UCE) are pretrained on tens of millions of cells drawn from public repositories. The same repositories underlie widely used integration benchmarks, creating an unmeasured risk that zero-shot benchmark performance reflects pretraining exposure rather than genuine generalization. We introduce \textbf{scContam}, a per-cell audit framework that combines a MinHash-based gene-set fingerprint signal against the explicit pretraining corpus with a loss-based membership inference attack (MIA-scFM). Applied to four scIB benchmarks and three scFMs, we find that two of the most-cited benchmarks, PBMC 3k and the CELLxGENE human pancreatic islet atlas, contain extensive pretraining-overlap evidence ($80.4\%$ and $77.0\%$ of cells with fingerprint $p < 0.05$ against Genecorpus-30M), whereas the post-cutoff datasets AIDA v2 and Tahoe-100M show no overlap evidence ($0\%$). A controlled re-pretraining experiment establishes that MIA-scFM AUROC scales monotonically with the model's capacity-to-data ratio (AUROC $0.494 \to 0.690 \to 0.881$ across properly-regularized, mildly-overfit, and aggressively-overfit regimes), demonstrating that production scFMs resist instance-level memorization but distributional contamination must be detected separately. A donor-matched, within-cell-type analysis with three architectures shows that contaminated cells embed measurably more tightly than donor-matched clean cells (permutation $p = 0.030, 0.014, < 0.002$, respectively), with a perfectly null AIDA negative control. Pretraining audits are tractable and should accompany scFM benchmark reporting.

q-bio.GN↗

arXiv2026

Critical Windows of Complexity Control: When Transformers Decide to Reason or Memorize

Recent work has shown that Transformers' compositional generalization is governed by \emph{complexity control}, initialization scale and weight decay, which steers training toward low-complexity reasoning solutions rather than high-complexity memorization. Existing analyses, however, treat complexity control as a single static hyperparameter choice, leaving open \emph{when} during training this control is actually decisive. We show that the memorization-versus-reasoning fate of a Transformer is determined within a sharp, identifiable window of training. On a controlled compositional task we find that (i)~weight decay applied for a single 25\%-of-training window matches full-training weight decay in out-of-distribution (OOD) accuracy ($0.93$ vs $0.91$); (ii)~holding total regularization budget constant, placing it in the middle of training yields $5{-}9\times$ higher OOD accuracy than placing it early; (iii)~the boundary of the critical window is remarkably sharp, window onset shifted by as little as $100$ optimization steps causes mean OOD to jump from chance ($0.15$) to reasoning-regime ($0.61$); (iv)~the window's position depends systematically on initialization scale, but the basin of attraction for reasoning solutions \emph{shrinks} at small initialization, contradicting the prevailing recommendation that smaller initialization is uniformly better. We further show that the critical-window phenomenon is task-specific: it does not appear on grokking with modular arithmetic, where properly tuned constant weight decay matches scheduled weight decay.

cs.LG↗

arXiv2026

Multi-Scale Reversible Chaos Game Representation: A Unified Framework for Sequence Classification

Biological classification with interpretability remains a challenging task. For this, we introduce a novel encoding framework, Multi-Scale Reversible Chaos Game Representation (MS-RCGR), that transforms biological sequences into multi-resolution geometric representations with guaranteed reversibility. Unlike traditional sequence encoding methods, MS-RCGR employs rational arithmetic and hierarchical k-mer decomposition to generate scale-invariant features that preserve complete sequence information while enabling diverse analytical approaches. Our framework bridges three distinct paradigms for sequence analysis: (1) traditional machine learning using extracted geometric features, (2) computer vision models operating on CGR-generated images, and (3) hybrid approaches combining protein language model embeddings with CGR features. Through comprehensive experiments on synthetic DNA and protein datasets encompassing seven distinct sequence classes, we demonstrate that MS-RCGR features consistently enhance classification performance across all paradigms. Notably, our hybrid approach combining pre-trained language model embeddings (ESM2, ProtT5) with MS-RCGR features achieves superior performance compared to either method alone. The reversibility property of our encoding ensures no information loss during transformation, while multi-scale analysis captures patterns ranging from individual nucleotides to complex motif structures. Our results indicate that MS-RCGR provides a flexible, interpretable, and high-performing foundation for biological sequence analysis.

cs.LG↗

arXiv2025

Uncovering Hierarchical Structure in LLM Embeddings with $\delta$-Hyperbolicity, Ultrametricity, and Neighbor Joining

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 $\delta$-hyperbolicity, Ultrametricity, and Neighbor Joining. $\delta$-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.

cs.CG↗

arXiv2025

DPSR: Differentially Private Sparse Reconstruction via Multi-Stage Denoising for Recommender Systems

Differential privacy (DP) has emerged as the gold standard for protecting user data in recommender systems, but existing privacy-preserving mechanisms face a fundamental challenge: the privacy-utility tradeoff inevitably degrades recommendation quality as privacy budgets tighten. We introduce DPSR (Differentially Private Sparse Reconstruction), a novel three-stage denoising framework that fundamentally addresses this limitation by exploiting the inherent structure of rating matrices -- sparsity, low-rank properties, and collaborative patterns. DPSR consists of three synergistic stages: (1) \textit{information-theoretic noise calibration} that adaptively reduces noise for high-information ratings, (2) \textit{collaborative filtering-based denoising} that leverages item-item similarities to remove privacy noise, and (3) \textit{low-rank matrix completion} that exploits latent structure for signal recovery. Critically, all denoising operations occur \textit{after} noise injection, preserving differential privacy through the post-processing immunity theorem while removing both privacy-induced and inherent data noise. Through extensive experiments on synthetic datasets with controlled ground truth, we demonstrate that DPSR achieves 5.57\% to 9.23\% RMSE improvement over state-of-the-art Laplace and Gaussian mechanisms across privacy budgets ranging from $\varepsilon=0.1$ to $\varepsilon=10.0$ (all improvements statistically significant with $p < 0.05$, most $p < 0.001$). Remarkably, at $\varepsilon=1.0$, DPSR achieves RMSE of 0.9823, \textit{outperforming even the non-private baseline} (1.0983), demonstrating that our denoising pipeline acts as an effective regularizer that removes data noise in addition to privacy noise.

cs.LG↗

arXiv2025

Boosting t-SNE Efficiency for Sequencing Data: Insights from Kernel Selection

Dimensionality reduction techniques are essential for visualizing and analyzing high-dimensional biological sequencing data. t-distributed Stochastic Neighbor Embedding (t-SNE) is widely used for this purpose, traditionally employing the Gaussian kernel to compute pairwise similarities. However, the Gaussian kernel's lack of data-dependence and computational overhead limit its scalability and effectiveness for categorical biological sequences. Recent work proposed the isolation kernel as an alternative, yet it may not optimally capture sequence similarities. In this study, we comprehensively evaluate nine different kernel functions for t-SNE applied to molecular sequences, using three embedding methods: One-Hot Encoding, Spike2Vec, and minimizers. Through both subjective visualization and objective metrics (including neighborhood preservation scores), we demonstrate that the cosine similarity kernel in general outperforms other kernels, including Gaussian and isolation kernels, achieving superior runtime efficiency and better preservation of pairwise distances in low-dimensional space. We further validate our findings through extensive classification and clustering experiments across six diverse biological datasets (Spike7k, Host, ShortRead, Rabies, Genome, and Breast Cancer), employing multiple machine learning algorithms and evaluation metrics. Our results show that kernel selection significantly impacts not only visualization quality but also downstream analytical tasks, with the cosine similarity kernel providing the most robust performance across different data types and embedding strategies, making it particularly suitable for large-scale biological sequence analysis.

cs.LG↗

arXiv2025

Robust Gradient Descent via Heavy-Ball Momentum with Predictive Extrapolation

Accelerated gradient methods like Nesterov's Accelerated Gradient (NAG) achieve faster convergence on well-conditioned problems but often diverge on ill-conditioned or non-convex landscapes due to aggressive momentum accumulation. We propose Heavy-Ball Synthetic Gradient Extrapolation (HB-SGE), a robust first-order method that combines heavy-ball momentum with predictive gradient extrapolation. Unlike classical momentum methods that accumulate historical gradients, HB-SGE estimates future gradient directions using local Taylor approximations, providing adaptive acceleration while maintaining stability. We prove convergence guarantees for strongly convex functions and demonstrate empirically that HB-SGE prevents divergence on problems where NAG and standard momentum fail. On ill-conditioned quadratics (condition number $\kappa=50$), HB-SGE converges in 119 iterations while both SGD and NAG diverge. On the non-convex Rosenbrock function, HB-SGE achieves convergence in 2,718 iterations where classical momentum methods diverge within 10 steps. While NAG remains faster on well-conditioned problems, HB-SGE provides a robust alternative with speedup over SGD across diverse landscapes, requiring only $O(d)$ memory overhead and the same hyperparameters as standard momentum.

cs.LG↗

arXiv2025

Sequence-to-Image Transformation for Sequence Classification Using Rips Complex Construction and Chaos Game Representation

Traditional feature engineering approaches for molecular sequence classification suffer from sparsity issues and computational complexity, while deep learning models often underperform on tabular biological data. This paper introduces a novel topological approach that transforms molecular sequences into images by combining Chaos Game Representation (CGR) with Rips complex construction from algebraic topology. Our method maps sequence elements to 2D coordinates via CGR, computes pairwise distances, and constructs Rips complexes to capture both local structural and global topological features. We provide formal guarantees on representation uniqueness, topological stability, and information preservation. Extensive experiments on anticancer peptide datasets demonstrate superior performance over vector-based, sequence language models, and existing image-based methods, achieving 86.8\% and 94.5\% accuracy on breast and lung cancer datasets, respectively. The topological representation preserves critical sequence information while enabling effective utilization of vision-based deep learning architectures for molecular sequence analysis.

cs.LG↗

arXiv2025

When do cheap embeddings beat protein language models? A theoretically-grounded hashing sketch for biological sequence classification

\textbf{Motivation:} Pre-trained protein language models (PLMs) such as ESM-2 have become the default representation for biological sequence tasks, but they are computationally heavy and require GPUs both for embedding and for fine-tuning. Whether they are actually necessary for sequence \emph{classification}, as opposed to structure prediction, is rarely tested against strong, principled, lightweight alternatives. This question has direct practical stakes for large-scale genomic surveillance, where embedding millions of sequences on commodity hardware is a recurring bottleneck.\\ \textbf{Results:} We introduce Murmur2Vec, an alignment-free, training-free embedding that aggregates $k$-mer counts into a small hash table via the deterministic MurmurHash function, and we cast it as a randomized sketch of the classical $k$-mer spectrum kernel. We provide a complete theoretical treatment: closed-form bias/variance of the inner product, an unbiased signed variant with a Johnson--Lindenstrauss-type concentration bound, an excess-risk bound for downstream linear classifiers that makes the bias--variance trade-off in the hash-table size explicit, and an implicit-regularization mechanism by which collisions damage frequent non-discriminative $k$-mers more than rare lineage-defining ones. Across four classification tasks, SARS-CoV-2 spike lineage (22 classes), HIV-1 Env subtype (8 classes), and two protein-family benchmarks (8 and 6 classes), Murmur2Vec matches a LoRA-fine-tuned 650M-parameter ESM-2 model on the two tasks for which LoRA fine-tuning was run to convergence (SARS-CoV-2 and HIV-1) and ties frozen ESM-2 on the two protein-family tasks, and it \emph{outperforms} the fine-tuned model on the hardest task (SARS-CoV-2 lineage: $0.854$ vs.\ $0.807$ accuracy; macro-F1 $0.684$ vs.\ $0.401$).

cs.LG↗

arXiv2025

Breaking the Euclidean Barrier: Hyperboloid-Based Biological Sequence Analysis

Genomic sequence analysis plays a crucial role in various scientific and medical domains. Traditional machine-learning approaches often struggle to capture the complex relationships and hierarchical structures of sequence data when working in high-dimensional Euclidean spaces. This limitation hinders accurate sequence classification and similarity measurement. To address these challenges, this research proposes a method to transform the feature representation of biological sequences into the hyperboloid space. By applying a transformation, the sequences are mapped onto the hyperboloid, preserving their inherent structural information. Once the sequences are represented in the hyperboloid space, a kernel matrix is computed based on the hyperboloid features. The kernel matrix captures the pairwise similarities between sequences, enabling more effective analysis of biological sequence relationships. This approach leverages the inner product of the hyperboloid feature vectors to measure the similarity between pairs of sequences. The experimental evaluation of the proposed approach demonstrates its efficacy in capturing important sequence correlations and improving classification accuracy.

cs.LG↗

arXiv2025

Explicit Path CGR: Maintaining Sequence Fidelity in Geometric Representations

We present a novel information-preserving Chaos Game Representation (CGR) method, also called Reverse-CGR (R-CGR), for biological sequence analysis that addresses the fundamental limitation of traditional CGR approaches - the loss of sequence information during geometric mapping. Our method introduces complete sequence recovery through explicit path encoding combined with rational arithmetic precision control, enabling perfect sequence reconstruction from stored geometric traces. Unlike purely geometric approaches, our reversibility is achieved through comprehensive path storage that maintains both positional and character information at each step. We demonstrate the effectiveness of R-CGR on biological sequence classification tasks, achieving competitive performance compared to traditional sequence-based methods while providing interpretable geometric visualizations. The approach generates feature-rich images suitable for deep learning while maintaining complete sequence information through explicit encoding, opening new avenues for interpretable bioinformatics analysis where both accuracy and sequence recovery are essential.

cs.LG↗

arXiv2025

Sequence Analysis Using the Bezier Curve

The analysis of sequences (e.g., protein, DNA, and SMILES string) is essential for disease diagnosis, biomaterial engineering, genetic engineering, and drug discovery domains. Conventional analytical methods focus on transforming sequences into numerical representations for applying machine learning/deep learning-based sequence characterization. However, their efficacy is constrained by the intrinsic nature of deep learning (DL) models, which tend to exhibit suboptimal performance when applied to tabular data. An alternative group of methodologies endeavors to convert biological sequences into image forms by applying the concept of Chaos Game Representation (CGR). However, a noteworthy drawback of these methods lies in their tendency to map individual elements of the sequence onto a relatively small subset of designated pixels within the generated image. The resulting sparse image representation may not adequately encapsulate the comprehensive sequence information, potentially resulting in suboptimal predictions. In this study, we introduce a novel approach to transform sequences into images using the B\'ezier curve concept for element mapping. Mapping the elements onto a curve enhances the sequence information representation in the respective images, hence yielding better DL-based classification performance. We employed different sequence datasets to validate our system by using different classification tasks, and the results illustrate that our B\'ezier curve method is able to achieve good performance for all the tasks.

q-bio.QM↗

arXiv2024

Hilbert Curve Based Molecular Sequence Analysis

Accurate molecular sequence analysis is a key task in the field of bioinformatics. To apply molecular sequence classification algorithms, we first need to generate the appropriate representations of the sequences. Traditional numeric sequence representation techniques are mostly based on sequence alignment that faces limitations in the form of lack of accuracy. Although several alignment-free techniques have also been introduced, their tabular data form results in low performance when used with Deep Learning (DL) models compared to the competitive performance observed in the case of image-based data. To find a solution to this problem and to make Deep Learning (DL) models function to their maximum potential while capturing the important spatial information in the sequence data, we propose a universal Hibert curve-based Chaos Game Representation (CGR) method. This method is a transformative function that involves a novel Alphabetic index mapping technique used in constructing Hilbert curve-based image representation from molecular sequences. Our method can be globally applied to any type of molecular sequence data. The Hilbert curve-based image representations can be used as input to sophisticated vision DL models for sequence classification. The proposed method shows promising results as it outperforms current state-of-the-art methods by achieving a high accuracy of $94.5$\% and an F1 score of $93.9\%$ when tested with the CNN model on the lung cancer dataset. This approach opens up a new horizon for exploring molecular sequence analysis using image classification methods.

cs.LG↗

arXiv2024

Converting Time Series Data to Numeric Representations Using Alphabetic Mapping and k-mer strategy

In the realm of data analysis and bioinformatics, representing time series data in a manner akin to biological sequences offers a novel approach to leverage sequence analysis techniques. Transforming time series signals into molecular sequence-type representations allows us to enhance pattern recognition by applying sophisticated sequence analysis techniques (e.g. $k$-mers based representation) developed in bioinformatics, uncovering hidden patterns and relationships in complex, non-linear time series data. This paper proposes a method to transform time series signals into biological/molecular sequence-type representations using a unique alphabetic mapping technique. By generating 26 ranges corresponding to the 26 letters of the English alphabet, each value within the time series is mapped to a specific character based on its range. This conversion facilitates the application of sequence analysis algorithms, typically used in bioinformatics, to analyze time series data. We demonstrate the effectiveness of this approach by converting real-world time series signals into character sequences and performing sequence classification. The resulting sequences can be utilized for various sequence-based analysis techniques, offering a new perspective on time series data representation and analysis.

cs.LG↗

arXiv2024

Computing Gram Matrix for SMILES Strings using RDKFingerprint and Sinkhorn-Knopp Algorithm

In molecular structure data, SMILES (Simplified Molecular Input Line Entry System) strings are used to analyze molecular structure design. Numerical feature representation of SMILES strings is a challenging task. This work proposes a kernel-based approach for encoding and analyzing molecular structures from SMILES strings. The proposed approach involves computing a kernel matrix using the Sinkhorn-Knopp algorithm while using kernel principal component analysis (PCA) for dimensionality reduction. The resulting low-dimensional embeddings are then used for classification and regression analysis. The kernel matrix is computed by converting the SMILES strings into molecular structures using the Morgan Fingerprint, which computes a fingerprint for each molecule. The distance matrix is computed using the pairwise kernels function. The Sinkhorn-Knopp algorithm is used to compute the final kernel matrix that satisfies the constraints of a probability distribution. This is achieved by iteratively adjusting the kernel matrix until the marginal distributions of the rows and columns match the desired marginal distributions. We provided a comprehensive empirical analysis of the proposed kernel method to evaluate its goodness with greater depth. The suggested method 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 several baseline methods in terms of supervised analysis and has potential uses in molecular design and drug discovery. Overall, the suggested method is a promising avenue for kernel methods-based molecular structure analysis and design.

cs.LG↗

arXiv2024

Neuromorphic Spiking Neural Network Based Classification of COVID-19 Spike Sequences

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.

cs.NE↗

arXiv2024

EPIC: Enhancing Privacy through Iterative Collaboration

Advancements in genomics technology lead to a rising volume of viral (e.g., SARS-CoV-2) sequence data, resulting in increased usage of machine learning (ML) in bioinformatics. Traditional ML techniques require centralized data collection and processing, posing challenges in realistic healthcare scenarios. Additionally, privacy, ownership, and stringent regulation issues exist when pooling medical data into centralized storage to train a powerful deep learning (DL) model. The Federated learning (FL) approach overcomes such issues by setting up a central aggregator server and a shared global model. It also facilitates data privacy by extracting knowledge while keeping the actual data private. This work proposes a cutting-edge Privacy enhancement through Iterative Collaboration (EPIC) architecture. The network is divided and distributed between local and centralized servers. We demonstrate the EPIC approach to resolve a supervised classification problem to estimate SARS-CoV-2 genomic sequence data lineage without explicitly transferring raw sequence data. We aim to create a universal decentralized optimization framework that allows various data holders to work together and converge to a single predictive model. The findings demonstrate that privacy-preserving strategies can be successfully used with aggregation approaches without materially altering the degree of learning convergence. Finally, we highlight a few potential issues and prospects for study in FL-based approaches to healthcare applications.

cs.LG↗