CS Paper Radar

Reducing double-dipping in single-cell topic models with three-way count splitting

Zheng Qu

cs.LG q-bio.GN ·submitted 2026-10-16 19:02 UTC · v1
code/dataunusual

Abstract

Topic models learn gene programs from single-cell RNA sequencing (scRNA-seq) data, and the same counts are then used to test which genes and donors are associated with those programs. Count splitting is the standard remedy for this reuse, but it has not been tested when an encoder reads each cell's counts. On a 104,400-cell lupus cohort with 300 permuted null genes per run, naive tests rejected 36% of null gene-topic pairs at a nominal 5% level, and the usual two-way split rejected 38%, because the encoder sees the tested gene's own held-out count. Encoding topic usage on a separate third of the counts, with beta-binomial thinning, lowered this to 7.7% (6.7% for genes outside the model); for two other models, LDVAE and scETM, the rate was 7.0-7.3% (naive 51-59%). A label-free interferon program shifts in held-out counts in the lupus cohort and in a pre-specified COVID-19 cohort, and donor-level topic usage predicts disease (AUROC 0.96) beyond cell-type proportions (0.86 and 0.58 in the two cohorts). Calibration and prediction did not depend on the model. We do not claim better clustering, prediction or topic stability. Code, figures and result tables are public (MIT license; DOI 10.5281/zenodo.23307193).

Comment
16 pages, 4 figures. Code: https://github.com/quz2004/topic-model-count-splitting

Closest papers

Nearest neighbours by meaning across the whole index.

When gene summaries reveal the answers: Gene Ontology circularity in gene embedding benchmarks

cs.LG q-bio.GN ·2026-11-08 ·Jiawei Wu, Huishu Wang, Junlin Chen, Hongyi Wang
similarity 0.50
Abstract · 31 pages, including supplementary material. GeneProv benchmark and leaderboard: https://hwang595.github.io/gene-embedding-exposure/

A gene summary can contain the very annotations used to evaluate an embedding of its text. The Alliance of Genome Resources converts Gene Ontology (GO) annotations into sentences, which NCBI Gene displays and GenePT includes in its input. These summaries covered 6,172 of 19,256 human protein-coding genes (32.1%), including 58% of genes with one to ten linked publications. We embedded the same corpus with and without the generated summaries and refitted a function-prediction probe on each version. Among genes with one to ten publications, removal reduced the reported per-gene area under the receiver operating characteristic curve (AUROC) by 0.078 with a linear probe (95% bootstrap interval 0.072-0.084) and by 0.033 with a multilayer perceptron. These estimates use a historical GO scorer that mishandled ties. The original predictions are unavailable for rescoring. A simple TF-IDF encoder also benefited from the summaries, as did models evaluated only on experimental annotations. Replacing generated summaries with UniProt function descriptions recovered much of the lost performance despite little verbatim overlap. UniProt also curates many of the evaluation labels. In mouse and yeast, we found a similar dependence on the source of the description. We introduce GeneProv to compare encoders on matched text and to examine how representation rankings change across GO recovery, non-GO endpoints and co-essentiality.

0.29

BarcodeMAE+: Rethinking Masked Pretraining and Global Representations for DNA Barcode Foundation Models

cs.LG q-bio.GN ·2026-09-26 ·Monireh Safari, Pablo Millan Arias, Scott C. Lowe, Lila Kari, Angel X. Chang +1 more
similarity 0.50
Abstract

Many DNA foundation models are pretrained by masking parts of a sequence and asking the model to reconstruct them. Standard masked pretraining exposes the encoder to special [MASK] tokens that are absent at inference, creating a mismatch between training and downstream use. The role of an explicit global sequence representation such as a [CLS] token and how it should be trained also remain poorly understood for DNA barcodes. We introduce BarcodeMAE+ and study model architecture, global [CLS] representation, and auxiliary pretraining objectives across arthropod COI (BIOSCAN-5M) and fungal ITS (UNITE+INSD) barcodes. Across both barcode regions, the encoder-decoder MAE-LM architecture outperforms its matched encoder-only counterpart in nearly all evaluated configurations, supporting MAE-LM as an effective architectural design for DNA barcode foundation models. A trained global [CLS] representation provides substantial additional gains: on BIOSCAN-5M, [CLS] accuracy increases from 47.53% without an auxiliary objective to 80.65% with cross-entropy genus classification. The best auxiliary objective is region-dependent: cross-entropy performs best on BIOSCAN-5M, whereas pairwise same-genus classification performs best on UNITE+INSD, reaching 73.19% on Yeast and 63.07% on Filamentous Fungi. BarcodeMAE+ outperforms published DNA foundation model baselines on BIOSCAN-5M and achieves the highest Yeast accuracy among the evaluated UNITE+INSD baselines using frozen encoder representations. Similarity-weighted softmax KNN voting further stabilizes accuracy as neighbourhood size increases. Overall, encoder-decoder masked pretraining and an explicitly trained global representation are strong design choices for DNA barcode foundation models, while the optimal objective for learning that representation depends on the biological domain.

0.25

Predictive Multiplicity in Cell-Fate Assignment: Label-Free Rashomon Sets and the Limits of Per-Cell Certification

cs.LG q-bio.GN ·2026-10-08 ·Arjun Bhupatiraju, Abhiram Bhupatiraju
similarity 0.50
Abstract · 14 pages

Single-cell trajectory inference maps transcriptomic measurements onto developmental continua, yet configurations that fit the data equally well can assign conflicting cell fates. FateMultiplicity is a label-free framework that constructs a statistically admissible model set, or Rashomon set, without lineage labels, by evaluating model discrepancy on cross-fitted held-out genes under non-inferiority testing calibrated against random-seed variation. Multiplicity is large and depends more on the diversity of the model space than its size: twelve configurations of a second algorithm expose 20.0% of cells where twenty-four of the first expose 3.8%. Whether the per-cell certified fate margin FM yields more reliable assignments than the fitted model already provides is then tested, and it does not. On simulation ground truth, on the same cells, FM discriminates misassignment at AUC 0.682, against 0.965 for the baseline configuration's own decision margin (p = 0.003) and 0.854 for a seed-dispersion baseline. Informativeness is governed by the breadth of the admitted set, not its cardinality: at cardinality four, seed refits give 0.933 and hyperparameter-perturbed sets 0.701. Relaxing the infimum to a q-quantile recovers discrimination but converges toward the single model's own confidence; the supremum reaches 0.973 because theta*'s membership bounds it from below, while the infimum is unanchored. Multiplicity in trajectory inference is worth measuring and reporting, but per-cell certification over a label-free Rashomon set is not a route to more reliable fate calls. Two constructions survive: a margin-erosion ratio separates real from spurious branch points in simulation (AUC 0.890, untested on real data), and against clonally observed fate, uncertified cells disagree with their clone's outcome 16.4 percentage points more often than certified cells (p < 0.001).

0.29

scAnyOmics: Foundation-Model Distillation for Any-to-Any Completion of Multi-omics Virtual Cells

cs.AI ·2026-10-30 ·Zhaoqi Song, Ye Du, Yusuke Iwasawa, Yutaka Matsuo, Qian Niu
similarity 0.49
Abstract · 14 pages, 4 figures, 2 tables

Recent advances in single-cell technologies have enabled the profiling of multiple molecular modalities, including transcriptomics, proteomics, and chromatin accessibility, providing a more comprehensive understanding of cellular states. However, integrating heterogeneous modalities and predicting missing molecular information remain challenging, particularly when paired measurements are limited. Here, we introduce scAnyOmics, a generative framework that distills knowledge from a frozen single-cell foundation model into a multimodal student model. scAnyOmics integrates scRNA-seq, protein-marker measurements, and scATAC-seq into a shared representation, enabling cross-modal prediction and reconstruction of missing modalities. We assembled 690,880 cells from six publicly available studies, covering both paired and partially paired multi-omics measurements, to systematically evaluate its predictive performance. We further evaluated scAnyOmics on held-out donor data, where it outperformed baseline methods across multiple evaluation metrics. Beyond cross-modal reconstruction, we explored its applications in cell-type annotation, perturbation-response prediction, and biological age estimation. Overall, scAnyOmics provides a unified framework for integrating incomplete single-cell multi-omics data and transferring pretrained biological knowledge, supporting the development of virtual cell models across diverse cellular contexts and experimental platforms.

0.22

Diversity Preservation Recovers Transcription Factor Motifs Across Challenging Confounder Landscapes

cs.NE q-bio.GN ·2026-11-16 ·Alejandro Medina, Mary Lauren Benton
similarity 0.49
Abstract

Motif discovery is a process where DNA sequences are searched for the short patterns, called motifs, that transcription factor (TF) proteins bind. However, the motif lengths are often unknown and motif attribution often competes with a near-identical motif of a related protein. In this work, we pair a motif genotype whose length emerges from an evolvable gate with MAP-Elites archives organized by three descriptor pairs, and validate every motif against the full HOCOMOCO database. On five liver ChIP-seq datasets spanning distinct confounder landscapes, a $(μ + λ)$ baseline differing only in survivor selection collapses onto a single motif and never ranks the target first across three datasets, while every MAP-Elites archive does so in at least 13 of 30 seeds on every dataset. Against the motif discovery tool MEME, run on the same sequences and validated identically, the archives recover the target signal as reliably on four datasets and more robustly when motif width is constrained. Neither method ranks the target above its closest relatives on HNF4A and JUND over the full width range, but narrow archive motifs rank HNF4A first in up to 43% of seeds, where MEME misses the motif in all settings. Diversity preservation improves our ability to recover consistent motif signal, although motif annotation remains a challenging problem for all methods we tested.

0.29

Leakage-aware calibrated decision-level fusion for mental-health text classification

cs.CL ·2026-11-14 ·Wadii Boulila, Giuseppe Varone, Bilel Benjdira, Maha Driss
published similarity 0.49
Abstract

Decision-level fusion benefits from complementary errors, calibrated probabilities, and leakage-aware evaluation. We evaluated a heterogeneous transformer fusion framework on 262,276 proxy-labeled Reddit records covering anxiety, depression, schizophrenia, and suicidal ideation. In a stratified random split, MinHash-LSH identified 29.1% of test records as lexical near-duplicates of training data. Augmentation-lineage auditing found 50.2% lineage overlap with training, including 30.0% of records retained after lexical filtering. We therefore used lineage-disjoint partitions. Twelve encoders, comprising three seeds each of Bio-ClinicalBERT, MentalBERT, MentalRoBERTa, and RoBERTa-base, underwent cross-fitted validation for calibration and ensemble selection. A class-specific Complementarity Index (CI), defined as the Jaccard distance between members' correctly classified validation sets, was assessed against independent fusion gain. CI correlated with gain for depression (r=0.50) and suicidal ideation (r=0.68), but not anxiety; results for schizophrenia were mixed. The selected MentalBERT, MentalRoBERTa, and RoBERTa-base triad achieved macro-F1 0.844 [0.823, 0.863], MCC 0.779, and ECE 0.0033 after post-fusion recalibration, improving macro-F1 by 0.021 [0.007, 0.037] over the validation-selected best single encoder. Integrated-Gradients attribution diversity was 2.3 times greater in fusion-rescued cases. Removing divergence-driving words destroyed 58% of rescues, versus 12% under random removal. These findings show that lineage auditing detects dependencies missed by lexical filtering, while class-specific complementarity and post-fusion calibration reveal ensemble behavior obscured by aggregate accuracy.

0.33

Different Models, Shared Failures: Auditing Frozen-Encoder Pools for Medical Segmentation

cs.CV cs.AI ·2026-10-30 ·Eunseob Choi, Kyeonghun Kim, Hyuk-Jae Lee, Nam-Joon Kim
top venuecode/datahot direction similarity 0.48
Abstract · Accepted at NeurIPS 2026, Track on Evaluations and Datasets. 47 pages, 8 figures. Code and data: https://github.com/eunseobchoi/CoFailMed

Average segmentation accuracy and model count do not reveal whether models fail on the same cases. We present CoFailMed, an audit of correlated failures in pretrained-encoder pools for medical segmentation. Across five 2D tasks, we evaluate 11 public pretrained encoders (a controlled sample, not a census) using frozen features and separately trained $1\times 1$ readouts with four seeds per encoder. Shared-checkpoint pairs have the highest mean per-case Dice correlation. An expanded 17-encoder analysis shows that distinct checkpoints within the same encoder family remain more correlated than cross-family pairs. At the pooled 33rd-percentile Dice threshold, four-member pools drawn from distinct encoders have mean effective sizes of 1.56-2.92, estimated from failure correlations in pools with enough failures and passes per member. In the primary pool, richer readouts and fine-tuning narrow the Dice-correlation gap between pairs sharing a checkpoint (or the DINOv2-B/S release) and all other pairs. Yet dependence between different encoders can increase, and fine-tuning does not consistently increase effective pool size at 33rd-percentile thresholds recomputed for each training recipe. A smaller between-group gap can coexist with greater dependence. Pool evaluation should therefore report absolute dependence and threshold-specific effective size alongside accuracy and nominal pool size. The CoFailMed release provides aligned per-case traces, analysis scripts and study code to reproduce the central audit results.

0.63

scTrilemma: Balancing Identity, Invariance, and Fidelity in Single-Cell Representation Learning

cs.LG cs.AI ·2026-09-30 ·Yunhak Oh, Yoonho Lee, Junseok Lee, Namkyeong Lee, Sang-Yeon Hwang +7 more
publishedcode/data similarity 0.48
Abstract · NeurIPS 2026

Single-cell RNA-seq representation learning is fundamentally label-free: cell identities, states, and contexts are not fixed training targets, so what constitutes signal or nuisance is analysis-dependent. A single representation must therefore preserve biological identity and state, remain robust to nuisance context, and retain the gene-level variation needed for expression analysis, three demands we call the representation trilemma. To tackle this problem, we introduce scTrilemma, a latent-bottleneck VAE that routes expression-derived variation to the embedding, the decoder, or the prior rather than forcing all of it through one embedding. It gates gene tokens by expression, routes the cell representation through the decoder, and conditions the prior on unlabeled pseudo-bulk context, under a single reconstruction objective and without target annotations or auxiliary representation losses. In release-based zero-shot evaluation on successive CZ CELLxGENE Census releases, scTrilemma leads all three demands at once and preserves biological-state, differential-expression, and pathway structure across multiple disease settings. Latent interventions further show that context can be removed at almost no cost to the other demands, leaving identity against fidelity as the remaining tension. Code is publicly available at https://github.com/yunhak0/scTrilemma.

0.48