Pacpaint: a histology-based deep learning model uncovers the extensive intratumor molecular heterogeneity of pancreatic adenocarcinoma.

Two tumor (Classical/Basal) and stroma (Inactive/active) subtypes of Pancreatic adenocarcinoma (PDAC) with prognostic and theragnostic implications have been described. These molecular subtypes were defined by RNAseq, a costly technique sensitive to sample quality and cellularity, not used in routine practice. To allow rapid PDAC molecular subtyping and study PDAC heterogeneity, we develop PACpAInt, a multi-step deep learning model. PACpAInt is trained on a multicentric cohort (n = 202) and validated on 4 independent cohorts including biopsies (surgical cohorts n = 148; 97; 126 / biopsy cohort n = 25), all with transcriptomic data (n = 598) to predict tumor tissue, tumor cells from stroma, and their transcriptomic molecular subtypes, either at the whole slide or tile level (112 µm squares). PACpAInt correctly predicts tumor subtypes at the whole slide level on surgical and biopsies specimens and independently predicts survival. PACpAInt highlights the presence of a minor aggressive Basal contingent that negatively impacts survival in 39% of RNA-defined classical cases. Tile-level analysis ( > 6 millions) redefines PDAC microheterogeneity showing codependencies in the distribution of tumor and stroma subtypes, and demonstrates that, in addition to the Classical and Basal tumors, there are Hybrid tumors that combine the latter subtypes, and Intermediate tumors that may represent a transition state during PDAC evolution.

A critical regulator of Bcl2 revealed by systematic transcript discovery of lncRNAs associated with T-cell differentiation.

Normal T-cell differentiation requires a complex regulatory network which supports a series of maturation steps, including lineage commitment, T-cell receptor (TCR) gene rearrangement, and thymic positive and negative selection. However, the underlying molecular mechanisms are difficult to assess due to limited T-cell models. Here we explore the use of the pro-T-cell line P5424 to study early T-cell differentiation. Stimulation of P5424 cells by the calcium ionophore ionomycin together with PMA resulted in gene regulation of T-cell differentiation and activation markers, partially mimicking the CD4-CD8- double negative (DN) to double positive (DP) transition and some aspects of subsequent T-cell maturation and activation. Global analysis of gene expression, along with kinetic experiments, revealed a significant association between the dynamic expression of coding genes and neighbor lncRNAs including many newly-discovered transcripts, thus suggesting potential co-regulation. CRISPR/Cas9-mediated genetic deletion of Robnr, an inducible lncRNA located downstream of the anti-apoptotic gene Bcl2, demonstrated a critical role of the Robnr locus in the induction of Bcl2. Thus, the pro-T-cell line P5424 is a powerful model system to characterize regulatory networks involved in early T-cell differentiation and maturation.

A comprehensive catalog of LncRNAs expressed in T-cell acute lymphoblastic leukemia.

Several studies have demonstrated that LncRNAs can play major roles in cancer development. The creation of a catalog of LncRNAs expressed in T cell acute lymphoblastic leukemia (T-ALL) is thus of particular importance. However, this task is challenging as LncRNA expression is highly restricted in time and space manner and thus may greatly differ between samples. We performed a systematic transcript discovery in RNA-Seq data obtained from T-ALL primary cells and cell lines. This led to the identification of 2560 novel LncRNAs. After the integration of these transcripts into a large compendium of LncRNAs (n = 30478) containing both known LncRNAs and those previously described in T-ALLs, we then performed a systematic genomic and epigenetic characterization of these transcript models demonstrating that these novel LncRNAs share properties with known LncRNAs. Finally, we provide evidence that these novel transcripts could be enriched in LncRNAs with potential oncogenic effects and identified a subset of LncRNAs coregulated with T-ALL oncogenes. Overall, our study represents a comprehensive resource of LncRNAs expressed in T-ALL and might provide new cues on the role of lncRNAs in this type of leukemia.