bims-fragic 2026-07-12 papers

Issue of 2026–07–12
four papers selected by
Laura Mannarino, Humanitas Research

Sci Adv. 2026 Jul 10. 12(28): eady9432

A scalable deep-learning framework for cancer detection using cell-free DNA shallow whole-genome sequencing.

Cell-free DNA (cfDNA) in body fluids enables noninvasive cancer detection. Multifeature artificial intelligence (AI) can improve sensitivity by integrating diverse biomarkers when cancer signals are sparse. Tumor-informed assays that rely on mutations have limited practicality for early cancer detection. Emerging fragmentomic and epigenetic features underpin tumor-naive approaches to screening for individuals with low tumor burden. Here, we designed UNITE-a universal cfDNA feature ensemble framework that provides scalable cancer detection methods based on "genomic bin-fragment length" matrices derived from shallow whole-genome sequencing (sWGS) data at 0.1× depth. Using sWGS data from 2063 plasma samples (631 controls and 1432 cases from 26 cancer types), we systematically evaluated both XGBoost (UNITE-XGB) and convolutional neural networks (UNITE-CNN) across multiple feature spaces and cancer stages. In stage I-II cancer, UNITE-XGB and UNITE-CNN achieved 31 and 21% sensitivity, respectively, at 95% specificity. These findings provide roadmaps for developing multifeature AI beyond plasma biopsies.

DOI: https://doi.org/10.1126/sciadv.ady9432

JHEP Rep. 2026 Jul 07. pii: S2589-5559(26)00226-0. [Epub ahead of print] 101955

A Pilot Study of Plasma Cell-Free DNA Fragmentomics in Gallbladder Cancer.

Michael Daniels, R Taylor Sundby, Ling Su, Xiaolin Wu, Ruth M Pfeiffer, Bin Zhu, Asif Rashid, Juan Carlos Araya, Juan Carlos Roa, Jonathan Hernandez, Catterina Ferreccio, Ann Hsing, Yu-Tang Gao, Jill Koshiol.

BACKGROUND & AIMS: Gallbladder cancer (GBC) is a rare and highly lethal biliary tract cancer with limited treatment options and lack of diagnostic and prognostic noninvasive biomarkers. Circulating cell-free DNA (cfDNA) offers a noninvasive means to capture fragmentomics alterations that may assist with diagnosis and biological characterization. This pilot study aimed to identify cfDNA-based features that differentiate GBC from individuals with gallstones and healthy controls.
METHODS: cfDNA was extracted from archived plasma samples from 67 individuals in two case-control studies from China and Chile, followed by low coverage whole genome sequencing to evaluate fragmentomics and related features. cfDNA computational packages were leveraged to generate features and create a classification model. External cfDNA datasets including other hepatopancreatobiliary disease groups were processed and compared to the current study.
RESULTS: In this pilot, individuals with GBC displayed significantly altered cfDNA features compared to healthy controls and individuals with gallstones (P<0.05). Key differences were observed in fragment lengths, end motif patterns, estimated tumor fractions, detectable copy number alterations, and transcription factor binding site accessibility, all of which discriminated GBC from a combined non-cancer group (AUC: 0.852). Many of these cfDNA alterations were consistent with the results from paired and unpaired tissue genomics datasets and other related cancer groups (liver cancer, pancreatic cancer, and non-GBC biliary tract cancer).
CONCLUSIONS: This proof-of-concept study demonstrates that archived plasma samples can be successfully used for cfDNA sequencing. These methods capture biologically meaningful alterations in GBC that are consistent with tissue-based genomics data. Collectively, these findings highlight the potential of cfDNA profiling for biological characterization and as a promising noninvasive diagnostic tool for GBC.
IMPACT AND IMPLICATIONS: This pilot study demonstrates that archived plasma EDTA samples can be used for cfDNA sequencing and reinforces the utility of cfDNA fragmentomics analyses for studying gallbladder disease. By assessing biologically relevant cfDNA features across related hepatopancreatobiliary cancers, we identified common and distinct features that may be used for classification and risk stratification. In high-risk settings for GBC, cfDNA fragmentomics might offer complementary information that could be used to guide clinical decision-making or help optimize waitlists for cholecystectomy.

Keywords: Cell-Free Nucleic Acids; Cholelithiasis; Gallbladder Neoplasm; Liquid Biopsy

DOI: https://doi.org/10.1016/j.jhepr.2026.101955

Brief Bioinform. 2026 Jul 03. pii: bbag359. [Epub ahead of print]27(4):

Exploring per-base quality scores as a surrogate marker of cell-free DNA fragmentome.

Hadas Volkov, Maria Raitses-Gurevich, Meitar Grad, Rani Shlayem, Dev Leibowitz, Tami Rubinek, Talia Golan, Noam Shomron.

Per-base quality scores are widely treated as technical metadata in next-generation sequencing. Here, we show that in rigorously controlled whole-genome sequencing of cell-free DNA, quality profiles may carry fragmentomic-associated signal that enables classification of cancer samples against matched controls. Analyzing four independent batches (23 cancer samples: pancreatic and breast; 22 matched controls) sequenced in a within-lane regime and further normalized per flow-cell tile to reduce technical confounders, we demonstrate through unsupervised analysis that boundary-enriched dynamics captured in these quality scores consistently separate cancer from control samples. A leave-one-batch-out classifier trained on quality-derived scores achieved a pooled area under the curve of 0.81. Furthermore, we show that the quality-derived metric correlates with short-fragment enrichment and tumor-associated 5'-end motifs, performing comparably to established, motif-based orthogonal methods. These results provide initial evidence that quality scores could serve as a low-cost, alignment-free surrogate signal for cfDNA-based cancer detection.

Keywords: NGS; biomarkers; cell-free DNA; fragmentomics; liquid biopsy; quality scores

DOI: https://doi.org/10.1093/bib/bbag359

Bioinformatics. 2026 Jul 01. pii: btag274. [Epub ahead of print]42(Supplement_1):

MOCDT: multi-cancer detection and tissue-of-origin classification via cfDNA multi-modal integration.

Gihyeon Kim, Seungyeon Rhee, Yumi Lee, Seongmun Jeong, Tae-You Kim, Jang-Hwan Choi.

MOTIVATION: Tumor-derived circulating tumor DNA (ctDNA) fragments present in blood provide rich molecular signals for identifying cancer and mapping its tissue of origin. However, leveraging these heterogeneous signals requires robust computational integration methods. Existing multi-modal approaches often fail to capture both inter-modality structure and inter-patient relationships, limiting their utility for robust cancer detection (CD) and fine-grained tissue-of-origin (TOO) classification.
RESULTS: We propose MOCDT, a cell-free DNA (cfDNA) multi-omics framework that follows a clinically aligned two-stage pipeline: high-specificity CD followed by conditional TOO classification. MOCDT combines (i) a supervised multi-modal autoencoder incorporating adversarial modality alignment and supervised contrastive geometry shaping, with (ii) a latent space patient similarity network and (iii) a residual Graph Convolutional Network for relational learning. Applied to a cfDNA cohort including healthy controls and eight cancer types, MOCDT achieved 95.74% specificity and 96.22% sensitivity for CD at a high-specificity operating point, and 75.2% Top1 and 91.06% Top3 accuracy for TOO classification. Latent attribution analysis showed that the model learns tissue-dependent latent features rather than relying on a single universal biomarker axis. Together, these results demonstrate that MOCDT enables accurate and interpretable cfDNA-based multi-omics integration, supporting clinically relevant liquid biopsy applications.
AVAILABILITY AND IMPLEMENTATION: Code and Dataset are available at https://github.com/Ewha-AI/MOCDT.

DOI: https://doi.org/10.1093/bioinformatics/btag274