iVar, an Interpretation-Oriented Tool to Manage the Update and Revision of Variant Annotation and Classification.

The rapid evolution of Next Generation Sequencing in clinical settings, and the resulting challenge of variant reinterpretation given the constantly updated information, require robust data management systems and organized approaches. In this paper, we present iVar: a freely available and highly customizable tool with a user-friendly web interface. It represents a platform for the unified management of variants identified by different sequencing technologies. iVar accepts variant call format (VCF) files and text annotation files and elaborates them, optimizing data organization and avoiding redundancies. Updated annotations can be periodically re-uploaded and associated with variants as historically tracked attributes, i.e., modifications can be recorded whenever an updated value is imported, thus keeping track of all changes. Data can be visualized through variant-centered and sample-centered interfaces. A customizable search function can be exploited to periodically check if pathogenicity-related data of a variant has changed over time. Patient recontacting ensuing from variant reinterpretation is made easier by iVar through the effective identification of all patients present in the database carrying a specific variant. We tested iVar by uploading 4171 VCF files and 1463 annotation files, obtaining a database of 4166 samples and 22,569 unique variants. iVar has proven to be a useful tool with good performance in terms of collecting and managing data from a medium-throughput laboratory.

Deep learning techniques for detecting preneoplastic and neoplastic lesions in human colorectal histological images.

Trained pathologists base colorectal cancer identification on the visual interpretation of microscope images. However, image labeling is not always straightforward and this repetitive task is prone to mistakes due to human distraction. Significant efforts are underway to develop informative tools to assist pathologists and decrease the burden and frequency of errors. The present study proposes a deep learning approach to recognize four different stages of cancerous tissue development, including normal mucosa, early preneoplastic lesion, adenoma and cancer. A dataset of human colon tissue images collected and labeled over a 10-year period by a team of pathologists was partitioned into three sets. These were used to train, validate and test the neural network, comprising several convolutional and a few linear layers. The approach used in the present study is 'direct'; it labels raw images and bypasses the segmentation step. An overall accuracy of >95% was achieved, with the majority of mislabeling referring to a near category. Tests on an external dataset with a different resolution yielded accuracies >80%. The present study demonstrated that the neural network, when properly trained, can provide fast, accurate and reproducible labeling for colon cancer images, with the potential to significantly improve the quality and speed of medical diagnoses.