Biology subjects

Lanciault, C.

Publications and source records attributed to Lanciault, C..

3 recordsLinked to original sources

bioRxiv2020

VISTA: Virtual ImmunoSTAining for pancreatic disease quantification in murine cohorts

Mechanistic studies of pancreatic disease progression using animal models require objective and quantifiable assessment of tissue changes among animal cohorts. Disease state quantification, however, relies heavily on tissue immunostaining, which can be expensive, labor- and time-intensive, and all too often produces uneven staining that is prone to variable interpretation between experts and inaccurate quantification. Here we develop a fully automated semantic segmentation tool using deep learning for the rapid and objective quantification of histologic features using hematoxylin and eosin (H&E) stained pancreatic tissue sections acquired from murine pancreatic cancer models. The tool was successfully trained to segment and quantify multiple histopathologic features of pancreatic pre-cancer evolution, including normal acinar structures, the ductal phenotype of acinar-to ductal metaplasia (ADM), dysplasia, and the expanding stromal compartment. Disease quantifications produced by our computational tool were highly correlated to the results obtained by immunostaining markers of normal and diseased tissue (DAPI, amylase, and cytokeratins; correlation score= 0.9, 0.95, and 0.91, respectively) and were able to accurately reproduce immunostain patterns. Moreover, our tool was able to distinguish ADM from dysplasia, which are not reliably distinguished by immunostaining, and avoid the pitfalls of uneven or poor-quality staining. Using this tool, we quantified the changes in histologic feature abundance for murine cohorts with oncogenic Kras-driven disease at 2 months and 5 months of age (n=12, n=13). The calculated changes in histologic feature abundance were consistent with biological expectations, showing an expansion of the stromal compartment, a reduction of normal acinar tissue, and an increase in both ADM and dysplasia as disease progresses (p= 2e-6, 6e-7, 4e-4, and 3e-5, respectively). These results demonstrate the tools efficacy for accurate and rapid quantification of multiple histologic features using an objective and automated platform. Our tool promises to rapidly accelerate and improve the quantification of altered pancreatic disease progression in animal studies.

bioinformatics↗

bioRxiv2019

SHIFT: speedy histological-to-immunofluorescent translation of whole slide images enabled by deep learning

Spatially-resolved molecular profiling by immunostaining tissue sections is a key feature in cancer diagnosis, subtyping, and treatment, where it complements routine histopathological evaluation by clarifying tumor phenotypes. In this work, we present a deep learning-based method called speedy histological-to-immunofluorescent translation (SHIFT) which takes histologic images of hematoxylin and eosin-stained tissue as input, then in near-real time returns inferred virtual immunofluorescence (IF) images that accurately depict the underlying distribution of phenotypes without requiring immunostaining of the tissue being tested. We show that deep learning-extracted feature representations of histological images can guide representative sample selection, which improves SHIFT generalizability. SHIFT could serve as an efficient preliminary, auxiliary, or substitute for IF by delivering multiplexed virtual IF images for a fraction of the cost and in a fraction of the time required by nascent multiplexed imaging technologies.\n\nKEY POINTSO_LISpatially-resolved molecular profiling is an essential complement to histopathological evaluation of cancer tissues.\nC_LIO_LIInformation obtained by immunofluorescence imaging is encoded by features in histological images.\nC_LIO_LISHIFT leverages previously unappreciated features in histological images to facilitate virtual immunofluorescence staining.\nC_LIO_LIFeature representations of images guide sample selection, improving model generalizability.\nC_LI

systems biology↗

bioRxiv2017

MicroRNA-451a regulates colorectal cancer radiosensitivity

Colorectal cancer (CRC) is a leading cause of cancer-related death. The responses of CRC to standard of care adjuvant therapies such as radiation or chemotherapy are poorly understood. MicroRNAs (miRs) are small non-coding RNAs that affect gene expression programs in cells by downregulating specific mRNAs. In this study, we discovered a set of microRNAs upregulated rapidly in response to a single 2 Gy dose fraction of {gamma}-radiation in a mouse colorectal carcinoma xenograft model. The most upregulated candidate in our signature, miR-451a inhibits tumor cell proliferation and attenuated surviving fraction in longer-term cultures. Conversely, inhibition of miR-451a increased proliferation, tumorsphere formation and surviving fraction of tumor cells. Using a bioinformatics approach, we identified four genes-CAB39, EMSY, MEX3C and EREG as targets of miR-451a. Transfection of miR-451a decreased both mRNA and protein levels of these targets. Importantly, we found miR-451a expression was decreased with tumor stage in a small subset of CRC patients. Finally, analysis of a TCGA colorectal cancer dataset reveals that the CAB39 and EMSY are upregulated at the protein level in a significant number of CRC patients and correlates with poorer overall survival. Taken together, our data indicates miR-451a influences the radiation sensitivity of colorectal carcinomas.

cancer biology↗