DeepFilter: A Deep Learning Based Variant Filter for VarDict

Abstract

With the development of sequencing technologies, somatic mutation analysis has become an important component in cancer research and treatment. VarDict is a commonly used somatic variant caller for this task.Although the heuristic-based VarDict algorithm exhibits high sensitivity and versatility, it may detect higher amounts of false positive variants than callers, limiting its clinical practicality. To address this problem, we propose DeepFilter, a deep-learning based filter for VarDict, which can filter out the false positive variants detected by VarDict effectively.Our approach trains two models for insertion-deletion mutations (InDels) and single nucleotide variants (SNVs), respectively. Experiments show that DeepFilter can filter at least 98.5% of false positive variants and retain 93.5% of true positive variants for InDels and SNVs in the commonly used tumor-normal paired mode. Source code and pre-trained models are available at https://github.com/LeiHaoa/DeepFilter.

Research on YOLOv5 Vehicle Detection and Positioning System Based on Binocular Vision

“…The YOLOv5 model uses the CIOU loss function by default to calculate the localization loss [22]. This function not only considers the overlap area between borders and the distance from the center point but also takes the aspect ratio into account when performing the bounding box regression calculation, which is shown in Equation (1). In the process of using the improved YOLOv5 algorithm to detect the target vehicle, the introduction of the CBAM attention mechanism can enhance the ability to pay attention to the vehicle area, help the model to pay more attention to the vehicle area and improve the performance of vehicle detection.…”

“…The rapid development of artificial intelligence has broadened the development space for intelligent transportation systems and autonomous driving technology [1]. In these applications, accurate vehicle detection and location is a crucial task, which is of great significance for realizing safe driving, improving traffic flow efficiency and improving the driving experience [2].…”