Airtime is crucial for high-rotation tricks in snowboard halfpipe performance, significantly impacting trick difficulty, the primary judging criterion. This study aims to enhance the detection of take-off and landing events using inertial measurement unit (IMU) data in conjunction with machine learning algorithms since manual video-based methods are too time-consuming. Eight elite German National Team snowboarders performed 626 halfpipe tricks, recorded by two IMUs at the lateral lower legs and a video camera. The IMU data, synchronized with video, were labeled manually and segmented for analysis. Utilizing a 1D U-Net convolutional neural network (CNN), we achieved superior performance in all of our experiments, establishing new benchmarks for this binary segmentation task. In our extensive experiments, we achieved an $80.34$% lower mean Hausdorff distance for unseen runs compared with the threshold approach when placed solely on the left lower leg. Using both left and right IMUs further improved performance ($83.37$% lower mean Hausdorff). For data from an algorithm-unknown athlete (Zero-Shot segmentation), the U-Net outperformed the threshold algorithm by $67.58$%, and fine-tuning on athlete-specific (Few-Shot segmentation) runs improved the lower mean Hausdorff to $78.68$%. The fine-tuned model detected takeoffs with median deviations of $0.008$ s (IQR $0.030$ s), landing deviations of $0.005$ s (IQR 0.020 s), and airtime deviations of $0.000$ s (IQR $0.027$ s). These advancements facilitate real-time feedback and detailed biomechanical analysis, enhancing performance and trick execution, particularly during critical events, such as take-off and landing, where precise time-domain localization is crucial for providing accurate feedback to coaches and athletes.

Purpose: A wide therapeutic repertoire has become available to oncologists including radio- and chemotherapy, small molecules and monoclonal antibodies. However, drug efficacy can be limited by genetic heterogeneity. Here, we designed a webtool that facilitates the data analysis of the in vitro drug sensitivity data on 265 approved compounds from the GDSC database in association with a plethora of genetic changes documented for 1001 cell lines in the CCLE data. Methods: The webtool computes odds ratios of drug resistance for a queried set of genetic alterations. It provides results on the efficacy of single compounds or groups of compounds assigned to cellular signaling pathways. Webtool availability: https://tools.hornlab.org/GDSC/. Results: We first replicated established associations of genetic driver mutations in BRAF, RAS genes and EGFR with drug response. We then tested the ‘BRCAness’ hypothesis and did not find increased sensitivity to the assayed PARP inhibitors. Analyzing specific PIK3CA mutations related to cancer and mendelian overgrowth, we found support for the described sensitivity of H1047 mutants to GSK690693 targeting the AKT pathway. Testing a co-mutated gene pair, GATA3 activation abolished PTEN-related sensitivity to PI3K/mTOR inhibition. Finally, the pharmacogenomic modifier ABCB1 was associated with olaparib resistance. Conclusions: This tool could identify potential drug candidates in the presence of custom sets of genetic changes and moreover, improve the understanding of signaling pathways. The underlying computer code can be adapted to larger drug response datasets to help structure and accommodate the increasingly large biomedical knowledge base.

A wide therapeutic repertoire has become available to oncologists including radio- and chemotherapy, small molecules and monoclonal antibodies. However, drug efficacy can be limited by genetic changes that allow cancer cells to escape therapy. Here, we designed a webtool that facilitates the data analysis of the Genomics of Drug Sensitivity in Cancer (GDSC) database on 265 approved compounds in association with a plethora of genetic changes documented for 1001 cell lines in the Cancer Cell Line Encyclopedia (CCLE, cBioPortal). The webtool computes odds ratios of drug resistance for a queried set of genetic alterations. It provides results on the efficacy of single or groups of compounds assigned to cellular signaling pathways. Using this webtool we replicated known genetic drivers and identified new candidate genes, germline variants, co-mutation, and pharmacogenomic modifiers of both cancer drug resistance and drug repurposing. Webtool availability: https://tools.hornlab.org/GDSC/ . Author Summary: Tumors develop through uncontrolled cell growth enabled by various alterations that create tumor heterogeneity. Changes of the genome and thus cancer cells cause every patient to react differently to drugs and can lead to drug resistance in cancer therapies. To overcome drug resistance, researchers focus on developing personalized therapies. Here, we provide a straightforward tool to test public in vitro drug sensitivity data on a range of drugs for custom analyses of genetic changes. This may inform the identification of potential drug candidates and improve our understanding of signaling pathways as we can test drug response with custom sets of genetic changes according to specific research questions. The tool and underlying code can be adapted to larger drug response datasets and other data types, e.g. metabolic data, to help structure and accommodate the increasingly large biomedical knowledge base.