Bennamoun, M. | bioRxiv Science

bioRxiv2026

Single-cell trajectories in metastatic urothelial carcinoma reveal tumor-immune reprogramming and macrophage-driven resistance to PD-(L)1 blockade

Immune checkpoint inhibitors (ICI) improved outcomes in metastatic urothelial carcinoma (mUC), but primary and acquired resistance remain poorly understood. We performed single-nuclei RNA sequencing on sequential metastatic biopsies from ICI-treated mUC patients. Tumor cells showed transcriptomic heterogeneity within individual lesions, basal cells being associated with increased immune infiltration and response. Myeloid and lymphoid compartments exhibited features of immune dysfunction in non-responders. Longitudinal analyses revealed convergent adaptive resistance mechanisms, dominated by polarization toward pro-tumoral macrophage states, but also including downregulation of the antigen presentation machinery in tumor cells, increased checkpoint expression with loss of cytotoxicity in T cells. Individual trajectories point to distinct evolutionary routes under ICI pressure. Across pivotal ICI trials, bulk expression of the M2-like macrophage marker HES1 predicted ICI resistance. Our study provides the first single-cell longitudinal atlas of ICI-treated mUC, revealing macrophage reprogramming as a dominant driver of resistance, establishing a framework for individualized immunotherapy strategies.

cancer biology↗

bioRxiv2022

DNABERT-based explainable lncRNA identification in plant genome assemblies

Long non-coding ribonucleic acids (lncRNAs) have been shown to play an important role in plant gene regulation, involving both epigenetic and transcript regulation. LncRNAs are transcripts longer than 200 nucleotides that are not translated into functional proteins but can be translated into small peptides. Machine learning models have predominantly used transcriptome data with manually defined features to detect lncRNAs, however, they often underrepresent the abundance of lncRNAs and can be biased in their detection. Here we present a study using Natural Language Processing (NLP) models to identify plant lncRNAs from genomic sequences rather than transcriptomic data. The NLP models were trained to predict lncRNAs for seven model and crop species (Zea mays, Arabidopsis thaliana, Brassica napus, Brassica oleracea, Brassica rapa, Glycine max and Oryza sativa) using publicly available genomic references. We demonstrated that lncRNAs can be accurately predicted from genomic sequences with the highest accuracy of 83.4% for Z. mays and the lowest accuracy of 57.9% for B. rapa, revealing that genome assembly quality might affect the accuracy of lncRNA identification. Furthermore, we demonstrated the potential of using NLP models for cross-species prediction with an average of 63.1% accuracy using target species not previously seen by the model. As more species are incorporated into the training datasets, we expect the accuracy to increase, becoming a more reliable tool for uncovering novel lncRNAs. Finally, we show that the models can be interpreted using explainable artificial intelligence to identify motifs important to lncRNA prediction and that these motifs frequently flanked the lncRNA sequence. HighlightsO_LIWe demonstrate for the first time the identification of lncRNAs from genomic sequences, instead of transcriptome sequences, allowing the identification of lowly expressed lncRNAs. C_LIO_LIA deep learning model (natural language processing) was employed to predict lncRNAs in two monocot and five dicot plant species. C_LIO_LIWe used explainable machine learning to extract the genomic motifs associated with lncRNA identification and highlighted potentially conserved structures. C_LI

bioinformatics↗