Patterns

Patterns

@cp-patterns.bsky.social

The science of data. A peer-reviewed #openaccess data science journal from @cellpress.bsky.social Editor-in-Chief: Andrew L Hufton (@alhufton.bsky.social) Visit us online at https://www.cell.com/patterns/

401 followers523 following249 posts

Patterns
@cp-patterns.bsky.social

Online Now: CytoVerse: Single-cell AI foundation models in the browser #datascience

dlvr.itCytoVerse: Single-cell AI foundation models in the browserCurrie et al. present CytoVerse, a framework that runs single-cell RNA-seq foundation models entirely in the web browser. Using ONNX model deployment and compressed IVFPQ indexing, it annotates local datasets against a 23-million-cell reference without server computation, installation, or data upload and shares embeddings as lightweight files, thereby enabling private, interactive, and collaborative single-cell analysis.
Patterns
@cp-patterns.bsky.social

Online Now: Estimating the accuracy of selective classifiers without labels using Venn-Abers predictors #datascience

dlvr.itEstimating the accuracy of selective classifiers without labels using Venn-Abers predictorsIn this work, the authors propose the use of Venn-Abers predictors to estimate the performance of selective classifiers when ground-truth labels are unavailable. Their approach provides calibrated interval estimates and practical lower bounds, helping quantify how reliably selective classifiers can operate when uncertain predictions are deferred for human review.
Patterns
@cp-patterns.bsky.social

Online Now: Robust inference via top-down feedback and neural variability in recurrent vision models #datascience

dlvr.itRobust inference via top-down feedback and neural variability in recurrent vision modelsGreco et al. show that top-down feedback improves the sensory robustness of recurrent vision models when combined with neural variability via dropout. Neural variability prevents unit-level co-adaptation, while feedback pathways stabilize population activity within compact low-dimensional representations. Their interaction suggests how biological visual systems may perform robust perceptual inference under uncertainty and offers a design principle for artificial vision systems that can generalize on out-of-distribution conditions.
Patterns
@cp-patterns.bsky.social

Online Now: Cross-species dendritic computation for biologically inspired decision-making #datascience

dlvr.itCross-species dendritic computation for biologically inspired decision-makingBiological dendrites process inputs through both linear and nonlinear mechanisms. Du et al. translate this principle into TNLP-LIF, a trainable three-compartment spiking neuron model. By combining cross-species working-memory simulations with reinforcement-learning tests, the study shows how dendritic integration can improve artificial neural networks while remaining a computational abstraction rather than a direct biological replica.
Patterns
@cp-patterns.bsky.social

Don't miss out -- this call for papers closes on November 2! www.cell.com/patterns/spe...

Patterns@cp-patterns.bsky.social

⚠ New Call for Papers ⚠ Patterns invites submission of Opinion papers from researchers and other experts working in low- and middle-income countries on topics relevant to data science, artificial intelligence and the broader data ecosystem Learn more: www.cell.com/patterns/spe...

Patterns
@cp-patterns.bsky.social

Online Now: Competition for attention predicts good-to-bad tipping in AI #datascience

dlvr.itCompetition for attention predicts good-to-bad tipping in AIWhen will your AI chat app—or your child’s, a doctor’s, a lawyer’s, or a soldier’s—tip unnoticed from good to bad output? Johnson and Huo derive a formula for this tipping point by using pen, paper, and high school vector math, focusing on how the AI pays attention. Predicted immediate vs. delayed tippings match well against open-weight AIs. It provides an AI control lever for how the tipping point depends on what the AI is told.
Patterns
@cp-patterns.bsky.social

Online Now: Large-scale CryoET segmentation from minimal expert annotations with foundation video models #datascience

dlvr.itLarge-scale CryoET segmentation from minimal expert annotations with foundation video modelsAnnotating 3D microscopy data is a major bottleneck in biomedical research. Schwartz et al. introduce SABER, a workflow that turns sparse expert notes on 2D slices into full 3D semantic segmentations. By pairing a simple classifier with a foundation segmentation model, SABER achieves dense supervision quality while cutting expert annotation time to less than an hour. This versatile tool scales up quantitative analysis across light and electron microscopy modalities.
Patterns
@cp-patterns.bsky.social

Online Now: A Bayesian framework for anomaly detection in scientific data based on Benford’s law #datascience

dlvr.itA Bayesian framework for anomaly detection in scientific data based on Benford’s lawSimulation-based Benfordness estimation (SBBE) quantifies how closely a dataset conforms to Benford’s law, the characteristic distribution of the first significant digits observed in many naturally occurring datasets and widely used in data auditing. Applicable across scientific domains, SBBE provides uncertainty estimates and enables meaningful comparisons across datasets of different sizes. When applied to bioactivity data, SBBE ranks databases by curation quality and flags anomalous subsets for which targeted expert review is likely to yield the greatest benefit.
Patterns
@cp-patterns.bsky.social

Online Now: Limitations of genomic language models for realistic sequence generation #datascience

dlvr.itLimitations of genomic language models for realistic sequence generationTzanakakis et al. benchmark long-form genomic sequences generated by Evo 2 and megaDNA against natural genomes across complementary compositional and organizational features. Synthetic sequences remain detectably distinct, diverge progressively from their conditioning seeds, and reveal that correcting one artifact does not ensure broader genomic realism.
Patterns
@cp-patterns.bsky.social

We are happy to be supporting an early-career researcher travel grant for the Banff International Research Station workshop on "Bridging Topology and Machine Learning for Data Science" sci.utah.edu/topology-and...

sci.utah.eduSCI Invites Grad Students and Postdocs to Apply for Topology and Machine-Learning Workshop Travel Grant - Scientific Computing and Imaging InstituteThanks to the generous sponsorship of Patterns and Cell Press, the University of Utah Scientific Computing and Imaging (SCI) Institute is pleased to offer the Patterns and Cell Press Early Career Trav...
Patterns
@cp-patterns.bsky.social

Online Now: Diversity in the impact of heterogeneities on recurrent networks performing a cognitive task #datascience

dlvr.itDiversity in the impact of heterogeneities on recurrent networks performing a cognitive taskNo two neurons in the brain are identical. In this study, Santhosh and Narayanan show that differences between neurons affect learning, task performance, activity patterns, and resistance to disruption in highly varied ways. Using many recurrent neural networks trained on the same cognitive task, they found that the effects depended on network hyperparameters, the amount and type of variation, and how performance was measured. Their results highlight the importance of studying neural networks as complete, interacting systems across many examples, rather than focusing on individual differences in isolation.
Patterns
@cp-patterns.bsky.social

Online Now: Unifying AI-assisted scientific discovery around exploration, hypothesis generation, and testing #datascience

dlvr.itUnifying AI-assisted scientific discovery around exploration, hypothesis generation, and testingIn this review, Hasib et al. introduce EXHYTE a process centric survey to unify research on AI-assisted scientific discovery. Mapping recent studies reveals rapid advances in evidence retrieval, knowledge assembly, and idea generation, but it also reveals a persistent bottleneck in translating hypotheses into executable tests and using results to refine subsequent steps. The survey underscores the need for human oversight, reproducibility, validation, and auditable end-to-end discovery workflows.
Patterns
@cp-patterns.bsky.social

Online Now: AI safety is a system property, not a sum of individual defenses #datascience

dlvr.itAI safety is a system property, not a sum of individual defensesSafety for large language models and artificial intelligence agents remains fragmented; defenses against jailbreaking coexist with unresolved hallucination, fairness, and autonomous-control problems. Chen et al. argue that these failures can interact and amplify each other, making safety a system property, rather than a collection of independent problems. They identify four safety challenge classes, ground their interactions in real-world incidents and a controlled clinical audit, and propose three design principles for integrated safety architectures.
Patterns
@cp-patterns.bsky.social

Online Now: Unifying statistical and mathematical modeling through a causal inference lens #datascience

dlvr.itUnifying statistical and mathematical modeling through a causal inference lensThis work proposes a conceptual framework to compare and contrast statistical and mathematical modeling using the notion of nonparametric partial identification from causal inference. The perspective adopted provides a set of assumptions that allow mathematical operations to be linked to causal claims, which are illustrated in a case study of pharmacodynamical modeling for a blood pressure medication.
Patterns
@cp-patterns.bsky.social

Online Now: Unsupervised random forests for interpretable representation analysis of protein conformational landscapes #datascience

dlvr.itUnsupervised random forests for interpretable representation analysis of protein conformational landscapesSahil et al. introduce unsupervised random forest (URF), a self-supervised feature-selection method that identifies informative structural descriptors from molecular dynamics simulations of proteins without requiring labels. A built-in learning coefficient lets URF assess its own training quality. Across ten protein systems, from folded enzymes to intrinsically disordered proteins, URF matches or outperforms supervised and deep learning baselines, recovers known functional residues, and improves downstream Markov state model construction, offering a general, interpretable, unsupervised framework for representation learning in biomolecular simulations.
Patterns
@cp-patterns.bsky.social

Online Now: Designing reproducible large-language-model-assisted scientific analyses #datascience

dlvr.itDesigning reproducible large-language-model-assisted scientific analysesResearchers are rapidly adopting large language models (LLMs) in their work, but this can create challenges for scientific reproducibility. In this tutorial, the authors offer an organizing question—does the LLM sit on or off the data path of the published analysis? They then explain how this question, combined with familiar goals of reproducibility, provenance, determinism, and interpretability, can help researchers deliberately reason about their AI use and preserve best practices.
Patterns
@cp-patterns.bsky.social

Online Now: Identifying low-quality features and suboptimal models to enhance phenotype prediction #datascience

dlvr.itIdentifying low-quality features and suboptimal models to enhance phenotype predictionShen et al. present mBoost, a diagnostic framework for building phenotype predictive models. mBoost evaluates model structure, feature redundancy, feature validity, and applicability to new data, helping researchers decide whether to use more complex models, do feature selection, or train a new model on new dataset.
Patterns
@cp-patterns.bsky.social

To help grow our benchmarks collection, we have released a new call for submissions that describe major, ground-breaking benchmarking studies or which present important benchmarking tools, platform or resources. Learn more: www.cell.com/patterns/spe...

Patterns@cp-patterns.bsky.social

New Benchmarks Collection We've launched a new collection showcasing our best benchmarking papers, including papers that present in-depth systematic comparisons as well as resources that advance benchmarking and push the field toward higher evidentiary standards. www.cell.com/patterns/col...

cell.comBenchmarks collection: PatternsExplore Patterns benchmarks studies through research articles, perspectives and expert opinions
Patterns
@cp-patterns.bsky.social

Online Now: The cost of reproducibility in artificial intelligence #datascience

dlvr.itThe cost of reproducibility in artificial intelligenceReproducibility is widely considered a cornerstone of the scientific method and thus a key element of research, yet researchers often experience a wide variety of issues when attempting to reproduce empirical studies in AI. This work explores the relationship between the reproducibility of publications and the standards of their corresponding venues.
Patterns
@cp-patterns.bsky.social

Online Now: Q-SID: Detecting collusion groups from exam question scores with error quantification #datascience

dlvr.itQ-SID: Detecting collusion groups from exam question scores with error quantificationOnline exams make coordinated answer sharing difficult to detect, while existing collusion-detection methods typically focus on suspicious pairs of students and are often restricted to multiple-choice responses. Yan et al. introduce Q-SID, which detects likely collusion groups, including groups larger than two, directly from graded question scores, and provides estimated false-positive rates. Across diverse real exams, Q-SID improves accuracy and speed, giving instructors a scalable screening tool for identifying cases that warrant careful follow-up review.
Patterns
@cp-patterns.bsky.social

Online Now: Re-analysis of CLAP supports prior evidence of PRC2 as an RNA-binding protein #datascience

dlvr.itRe-analysis of CLAP supports prior evidence of PRC2 as an RNA-binding proteinYongWoo Lee and colleagues re-examine raw datasets from a recent crosslinking heat-activated purification (CLAP) study on the polycomb repressive complex 2 (PRC2). Their findings reveal that non-standard computational preprocessing suppressed vital signals, and correcting these workflow steps successfully reaffirms PRC2 as an RNA-binding protein.

Patterns reposted

Charlie Rahal
@crahal.com

How reliable is research 🤔? We publish RobustiPy in @cp-patterns.bsky.social: a highly accessible, #GNU-GPL 3.0 #python lib for #robustness analysis and #datascience. It may represent the “next generation” of #modeluncertainty, bringing a range of frontier #statistics methods into one toolkit 📊📈!

Patterns
@cp-patterns.bsky.social

Online Now: Retrieval-augmented reasoning reshapes collective reliability in text-based radiology question answering #datascience

dlvr.itRetrieval-augmented reasoning reshapes collective reliability in text-based radiology question answeringCan AI systems be trusted when the model behind them keeps changing? Farajiamiri et al. evaluate 34 language models on text-based radiology questions and show that giving every model the same structured evidence makes their answers more consistent and more reproducibly correct. The gains come from synthesizing evidence, not retrieval alone, yet models still sometimes agree on the same wrong answer support.
Patterns
@cp-patterns.bsky.social

Online Now: Disentangling batch effects and biological signals across conditions in single-cell transcriptome #datascience

dlvr.itDisentangling batch effects and biological signals across conditions in single-cell transcriptomeBatch effects can obscure biological differences in single-cell transcriptomic studies, particularly when experimental batches are confounded with disease, treatment, or developmental stages. Sakaguchi et al. introduce Kanade, a variational autoencoder that separates batch effects from condition-associated signals. Kanade selectively removes batch effects while preserving biologically meaningful differences between conditions and produces corrected gene-expression values suitable for downstream analyses, thereby enabling more reliable comparisons across complex single-cell datasets.
Patterns
@cp-patterns.bsky.social

Online Now: BiomarkerKB: An integrated knowledgebase supporting biomarker-centric exploration of biomedical data #datascience

dlvr.itBiomarkerKB: An integrated knowledgebase supporting biomarker-centric exploration of biomedical dataBiomarkerKB addresses the fragmentation of biomarker information by integrating diverse biomarker resources into a standardized, computable knowledgebase. The resource combines over 200,000 biomarker-disease associations with rich contextual metadata and a knowledge graph accessible through a public web portal. BiomarkerKB supports data reuse, knowledge discovery, and the development of biomarker-driven research.
Patterns
@cp-patterns.bsky.social

Online Now: EmmaEmb: A quantitative framework for analyzing embedding spaces in molecular biology #datascience

dlvr.itEmmaEmb: A quantitative framework for analyzing embedding spaces in molecular biologyFoundation models generate embeddings that capture complex biological information, but interpreting and comparing these representations remains challenging. Rissom et al. introduce EmmaEmb, a model-agnostic framework for the analysis of embedding spaces. Applied to protein language models and high-stakes molecular biology and drug-discovery tasks, EmmaEmb ranks representations by task relevance, reveals patterns underlying prediction errors, and tracks representational changes during model adaptation, providing a quantitative approach to understanding learned biological representations. All methods are available as an open-source Python library.
Patterns
@cp-patterns.bsky.social

Online Now: Hybrid artificial intelligence and quantum annealing as an optimization layer in drug discovery #datascience

dlvr.itHybrid artificial intelligence and quantum annealing as an optimization layer in drug discoveryArtificial intelligence has significantly expanded candidate generation in drug discovery, but selecting optimal compounds under real-world experimental constraints remains a major bottleneck. This review examines quantum annealing as a potential downstream optimization layer within hybrid computational workflows. By focusing on constraint-dominated decision tasks formulated as quadratic unconstrained binary optimization (QUBO) problems, the authors analyze how annealing architectures complement AI-generated scores. They further discuss practical hardware limitations, reporting considerations, and the critical role of rigorous benchmarking against classical baselines.
Patterns
@cp-patterns.bsky.social

Online Now: Uncovering smooth structures in single-cell data with neighbor embeddings guided by predictability-computability-stability #datascience

dlvr.itUncovering smooth structures in single-cell data with neighbor embeddings guided by predictability-computability-stabilitySingle-cell data often capture continuous biological processes, yet common visualization tools can introduce misleading structure. This study systematically evaluates neighbor embedding methods using the predictability-computability-stability (PCS) framework, revealing key limitations in reliability. The authors introduce NESS, a stability-guided approach that improves the recovery of smooth biological patterns, enabling more trustworthy interpretation of cell-state dynamics across diverse datasets.
Patterns
@cp-patterns.bsky.social

Online Now: Revealing hidden material composition via physics-induced deep learning for precision clinical diagnosis #datascience

dlvr.itRevealing hidden material composition via physics-induced deep learning for precision clinical diagnosisLiu et al. present a physics-induced deep-learning approach that extracts tissue composition information from routine single-energy CT scans. By incorporating X-ray imaging principles into model design, this approach improves material characterization and enhances performance across diverse clinical tasks. The framework expands the diagnostic value of standard CT imaging and supports more accessible precision medicine.
Patterns
@cp-patterns.bsky.social

Online Now: Network-based disruption analysis of biological coordination #datascience

dlvr.itNetwork-based disruption analysis of biological coordinationBiological systems rely on coordinated interaction networks between molecules and cells that can be disrupted by physiological challenges. NeDis is an open-source framework that measures how these networks change. By identifying groups of biomarkers with characteristic disruption patterns across conditions and time points, NeDis reveals system-level adaptations that are missed by standard analyses. Applications to immune data uncover coordinated disruptions during pregnancy and preeclampsia.
Patterns
@cp-patterns.bsky.social

Online Now: Sampling bias corrections for discrete and Gaussian partial information decompositions #datascience

dlvr.itSampling bias corrections for discrete and Gaussian partial information decompositionsLorenz, Engel et al. characterize the limited-sampling bias in partial information decomposition calculations. The bias is substantial when computed on empirical brain datasets, but the bias-correction methods developed here are demonstrated to work highly effectively to produce unbiased estimates of information synergy and redundancy.
Patterns
@cp-patterns.bsky.social

Online Now: Interpretable machine learning reveals hemispheric asymmetry of state switching in the suprachiasmatic nucleus #datascience

dlvr.itInterpretable machine learning reveals hemispheric asymmetry of state switching in the suprachiasmatic nucleusZhang et al. developed an interpretable machine learning framework to analyze population-level calcium ion signals in the suprachiasmatic nucleus (SCN), the central circadian clock of mammals. They uncover a hemispheric asymmetry in the SCN, particularly during the subjective day. Time prediction experiments further show that each hemisphere encodes differential time-encoding features. The work highlights hemispheric asymmetry as a fundamental property of the brain’s circadian clock.
Patterns
@cp-patterns.bsky.social

Online Now: mlr3mbo: Bayesian optimization in R #datascience

dlvr.itmlr3mbo: Bayesian optimization in RMany problems in engineering, the sciences, and machine learning require optimizing a system with expensive evaluations. The authors present mlr3mbo, a rich R toolbox for Bayesian optimization (BO)—the standard methodology for such problems—that pairs a modular design of BO building blocks with well-performing, empirically chosen defaults and convenient entry points. On standard benchmarks, it performs on par with or better than HEBO, SMAC3, Ax, and Optuna.
Patterns
@cp-patterns.bsky.social

Online Now: A tropical geometry for bounded biochemical state spaces #datascience

dlvr.itA tropical geometry for bounded biochemical state spacesMany biological measurements are bounded: a protein site cannot be less than unmodified or more than fully modified. Cobley shows that treating such data with ordinary linear tools can hide biological structure. Using cysteine oxidation as an example, Oxi-Shapes applies tropical geometry to reveal how aging reorganizes the redox proteome despite small average changes, offering a broader framework for analyzing bounded biochemical state spaces.
Patterns
@cp-patterns.bsky.social

Online Now: OscillomeR infers ultradian oscillations and targets of the Hes family #datascience

dlvr.itOscillomeR infers ultradian oscillations and targets of the Hes familyOscillomeR is a master-transcription-factor-anchored framework that infers ultradian gene oscillations from bulk and single-cell RNA sequencing. By reconstructing cyclic trajectories and phase-resolved gene expression, the method recovers Hes-driven oscillations in asynchronous muscle stem cells and pancreatic progenitors. It maps widespread rhythmic programs and phase shifts, revealing dynamic rewiring of gene-regulatory networks and helping to dissect the biological functions of gene-expression oscillations across development and disease.
Patterns
@cp-patterns.bsky.social

Online Now: MedicalAgentsBench for complex medical reasoning: Comparing internalized reasoning models versus externalized agent-based frameworks #datascience

dlvr.itMedicalAgentsBench for complex medical reasoning: Comparing internalized reasoning models versus externalized agent-based frameworksShao et al. introduce MedicalAgentsBench, a benchmark designed to separate genuine multi-step clinical reasoning from pattern matching. Using this benchmark, they compare internalized reasoning models against externalized agent frameworks and find that the two approaches are complementary; layering agents onto an internalized reasoning model achieves the highest accuracy, exceeding either approach alone.
Patterns
@cp-patterns.bsky.social

Online Now: BioMaster: Multi-agent system for automated bioinformatics analysis workflow #datascience

dlvr.itBioMaster: Multi-agent system for automated bioinformatics analysis workflowBioMaster is a knowledge-guided multi-agent framework that automates bioinformatics workflow planning, execution, debugging, and validation. Across 49 tasks and 102 tools, it improves workflow completion over existing automated systems and supports both proprietary and open-source language models.
Patterns
@cp-patterns.bsky.social

Online Now: Efficient attention mechanisms for large language models #datascience

dlvr.itEfficient attention mechanisms for large language modelsThis survey reviews efficient attention mechanisms for scalable long-context language modeling, with a focus on linear and sparse attention. It summarizes their algorithmic foundations, hardware considerations, and integration into large-scale pre-trained language models, providing a systematic reference for designing efficient transformer-based architectures.