Coppens, L. | bioRxiv Science

bioRxiv2025

Comparative characterization of OncoPro and Wnt-Based media reveals distinct phenotypic and pharmacologic states in patient-derived tumor organoids

BackgroundPatient-derived tumor organoids (PDTOs) are strongly influenced by culture medium. We compared OncoPro (OP) Tumoroid Culture medium with conventional Wnt/R-spondin/noggin (Wnt) medium. This recently developed OP medium offers a standardized, serum-free alternative to Wnt-based formulations. MethodsWe compared OP and Wnt media across 36 PDTO lines from various malignancies (colorectal, pancreatic, breast, lung, gastric, gastroesophageal junction, biliary head- and neck and uknown primary), assessing establishment success. Selected PDTO models were subjected to downstream characterization, including morphological assessment and bulk transcriptomic profiling with comparison to public single-cell RNA sequencing reference datasets (n=11), whole-exome sequencing (WES) (n=9), and pharmacological response profiling to a 33-drug panel (n=3). ResultsAdaptation from Wnt medium to OP succeeded in 83.3% (15/18), whereas de novo establishment favored Wnt (33.3% vs 11.1%). Key oncogenic driver alterations were retained across matched organoid cultures, supporting preservation of tumor-relevant genomic features. Transcriptomic profiling confirmed preserved tumor-identity across media, while revealing different epithelial state programs: Wnt upregulated proliferation/stemness-associated genes (e.g. LGR5) and OP enriched adhesion-associated genes and inflammatory/TGF-{beta} programs. In scRNA databases OP signatures preferentially mapped to malignant epithelial compartments in pancreatic cancer, whereas Wnt signatures were linked to non-malignant epithelium. Similarly, in colon cancer OP signature mapped predominantly to the malignant epithelial compartments. Drug (n=33) screening in pancreatic- and colorectal cancer PDTOs (n=3) demonstrated consistent medium-dependent shifts: Wnt-grown PDTOs were globally more sensitive in the screened subset, particularly to MAPK-axis inhibitors and apoptosis-sensitizers, while OP-grown PDTOs exhibited relative resistance. ConclusionsCulture medium composition is a key determinant of PDTO phenotype, transcriptome and drug sensitivity. Wnt medium was associated with drug-sensitive states, whereas OP medium was associated with adhesion- and inflammatory-related programs, relative resistance in the screened subset of 33 drugs and closer alignment with malignant epithelial programs in the analyzed pancreatic / colorectal cancer single-cell atlases.

cancer biology↗

bioRxiv2025

PromoterAtlas: decoding regulatory sequences across Gammaproteobacteria using a transformer model

Recent advances in deep learning, particularly transformer architectures, have improved computational approaches for biological sequence analysis. Despite these advances, computational models for bacterial promoter prediction have remained limited by small datasets, species-specific training, and binary classification approaches rather than comprehensive annotation frameworks. We present PromoterAtlas, a 1.8M parameter transformer model trained on 9M regulatory sequences from 3,371 gammaproteobacterial species. The model demonstrates recognition of various regulatory elements across different species, including ribosomal binding sites, various types of bacterial promoters, transcription factor binding sites, and terminators. Using this model, we developed a whole-genome promoter annotation tool for Gammaproteobacteria, with various levels of validation that support the predictions of promoters associated with different sigma ({sigma}) factors. Furthermore, we show that the model embeddings encode cross-species evolutionary relationships, clustering promoters by {sigma} factor identity rather than species-specific sequence features. Finally, we show that model embeddings encode regulatory sequence information that enables effective prediction of transcription and translation levels. PromoterAtlas can contribute to our understanding of and ability to engineer bacterial regulatory sequences with potential applications in bacterial biology, synthetic biology, and comparative genomics.

bioinformatics↗