Supplemental Figure S4. Sample partitioning for training a MYOD1 mutation predictive model using K-fold cross-validation.

Supplemental Figure S3. Sample partitioning for training and testing a RAS pathway mutation predictive model using K-fold cross-validation. Three independent experiments were trained on a random selection of samples for training, validation and testing.

Supplemental Figure S6. Graphical User Interface for tissue segmentation, MYOD1 mutation prediction, and risk prediction models.

Supplemental Figure S5. Frequency of mutations in COG and A.I. designated risk groups.

Supplemental Figure S2. Sample partitioning for training and testing a TP53 mutation predictive model using K-fold cross-validation.

Supplemental Table S1. Whole slide image tissue segmentation statistics by an expert pathologist and probability prediction using a trained convolutional neural network. Supplemental Table S2. Clinical and molecular characteristics of FN-RMS samples used for training models for mutation prediction. Yellow boxes indicate genes included in defining the RAS pathway. Supplemental Table S3. Clinical information with COG risk stratification of FN-RMS samples used for training a prognostication predictive CNN.

Abstract Purpose: Rhabdomyosarcoma (RMS) is an aggressive soft-tissue sarcoma, which primarily occurs in children and young adults. We previously reported specific genomic alterations in RMS, which strongly correlated with survival; however, predicting these mutations or high-risk disease at diagnosis remains a significant challenge. In this study, we utilized convolutional neural networks (CNN) to learn histologic features associated with driver mutations and outcome using hematoxylin and eosin (H&E) images of RMS. Experimental Design: Digital whole slide H&E images were collected from clinically annotated diagnostic tumor samples from 321 patients with RMS enrolled in Children's Oncology Group (COG) trials (1998–2017). Patches were extracted and fed into deep learning CNNs to learn features associated with mutations and relative event-free survival risk. The performance of the trained models was evaluated against independent test sample data (n = 136) or holdout test data. Results: The trained CNN could accurately classify alveolar RMS, a high-risk subtype associated with PAX3/7-FOXO1 fusion genes, with an ROC of 0.85 on an independent test dataset. CNN models trained on mutationally-annotated samples identified tumors with RAS pathway with a ROC of 0.67, and high-risk mutations in MYOD1 or TP53 with a ROC of 0.97 and 0.63, respectively. Remarkably, CNN models were superior in predicting event-free and overall survival compared with current molecular-clinical risk stratification. Conclusions: This study demonstrates that high-risk features, including those associated with certain mutations, can be readily identified at diagnosis using deep learning. CNNs are a powerful tool for diagnostic and prognostic prediction of rhabdomyosarcoma, which will be tested in prospective COG clinical trials.

Natural materials' inherently weak nonlinear response demands the design of artificial substitutes to avoid optically large samples and complex phase-matching techniques. Silicon photonic crystals are promising artificial materials for this quest. Their nonlinear properties can be modulated optically, paving the way for applications ranging from ultrafast information processing to quantum technologies. A two-dimensional 15-μm-thick silicon photonic structure, comprising a hexagonal array of air holes traversing the slab's thickness, has been designed to support a guided resonance for the light with a wavelength of 4-μm. At the resonance conditions, a transverse mode of the light is strongly confined between the holes in the "veins" of the silicon component. Owing to the confinement, the structure exhibits a ratio of nonlinear to linear absorption coefficients threefold higher than the uniform silicon slab of the same thickness. A customised time-resolved Z-scan method with provisions to accommodate ultrafast pump-probe measurements was used to investigate and quantify the non-linear response. We show that optically pumping free charge carriers into the structure decouples the incoming light from the resonance and reduces the non-linear response. The time-resolved measurements suggest that the decoupling is a relatively long-lived effect on the scale comparable to the non-radiative recombination in the bulk material. Moreover, we demonstrate that the excited free carriers are not the source of the nonlinearity, as this property is determined by the structure design.

BackgroundApproximately 4-8% of the world suffers from a rare disease. Rare diseases are often difficult to diagnose, and many do not have approved therapies. Genetic sequencing has the potential to shorten the current diagnostic process, increase mechanistic understanding, and facilitate research on therapeutic approaches but is limited by the difficulty of novel variant pathogenicity interpretation and the communication of known causative variants. It is unknown how many published rare disease variants are currently accessible in the public domain.ResultsThis study investigated the translation of knowledge of variants reported in published manuscripts to publicly accessible variant databases. Variants, symptoms, biochemical assay results, and protein function from literature on the SLC6A8 gene associated with X-linked Creatine Transporter Deficiency (CTD) were curated and reported as a highly annotated dataset of variants with clinical context and functional details. Variants were harmonized, their availability in existing variant databases was analyzed and pathogenicity assignments were compared with impact algorithm predictions. 24% of the pathogenic variants found in PubMed articles were not captured in any database used in this analysis while only 65% of the published variants received an accurate pathogenicity prediction from at least one impact prediction algorithm.ConclusionsDespite being published in the literature, pathogenicity data on patient variants may remain inaccessible for genetic diagnosis, therapeutic target identification, mechanistic understanding, or hypothesis generation. Clinical and functional details presented in the literature are important to make pathogenicity assessments. Impact predictions remain imperfect but are improving, especially for single nucleotide exonic variants, however such predictions are less accurate or unavailable for intronic and multi-nucleotide variants. Developing text mining workflows that use natural language processing for identifying diseases, genes and variants, along with impact prediction algorithms and integrating with details on clinical phenotypes and functional assessments might be a promising approach to scale literature mining of variants and assigning correct pathogenicity. The curated variants list created by this effort includes context details to improve any such efforts on variant curation for rare diseases.

A range of the distinctive physical properties, comprising high surface-to-volume ratio, possibility to achieve mechanical and chemical stability after a tailored treatment, controlled quantum confinement and the room-temperature photoluminescence, combined with mass production capabilities offer porous silicon unmatched capabilities required for the development of electro-optical devices. Yet, the mechanism of the charge carrier dynamics remains poorly controlled and understood. In particular, non-radiative recombination, often the main process of the excited carrier's decay, has not been adequately comprehended to this day. Here we show, that the recombination mechanism critically depends on the composition of surface passivation. That is, hydrogen passivated material exhibits Shockley-Read-Hall type of decay, while for oxidised surfaces, it proceeds by two orders of magnitude faster and exclusively through the Auger process. Moreover, it is possible to control the source of recombination in the same sample by applying a cyclic sequence of hydrogenation-oxidation-hydrogenation processes, and, consequently switching on-demand between Shockley-Read-Hall and Auger recombinations. Remarkably, irregardless of the recombination mechanism, the rate constant scales inversely with the average volume of individual silicon nanocrystals contained in the material. Thus, the type of the non-radiative recombination is established by the composition of the passivation, while its rate depends on the degree of the charge carriers' quantum confinement.

A patient with profound respiratory distress was brought by ambulance to the emergency department. The patient was severely hypoxic and attempts to oxygenate him failed. He deteriorated rapidly, becoming unresponsive. Immediate intubation was necessary however a cannot intubate, cannot oxygenate situation evolved, followed by cardiac arrest. An anaesthetist performed an emergency Front of Neck Access (eFONA) procedure which was successful. The cause of airway obstruction in this case was acute epiglottitis. Scalpel cricothyroidotomy can be a lifesaving procedure when performed in a timely manner, even for a patient with hypoxic cardiac arrest.

Background: Rhabdomyosarcoma (RMS) is an aggressive soft tissue tumor in children and young adults, accounting for 350-400 new cases annually in the US. Diagnosis of RMS is defined by the expression of genes related to skeletal muscle differentiation and can be further subclassified based on histological patterns (embryonal, ERMS; alveolar, ARMS; spindle/sclerosing, SSRMS). Genetic studies have found that the presence of a PAX fusion gene (FP-RMS), which is present in many ARMS tumors, correlates with poor outcome. Subsequent studies have identified additional genetic alterations (ex. TP53 or MYOD1 mutations) which also display distinct histological features and are associated with poor outcome. As a result, there is a growing need to identify these mutations to improve risk stratification. The goal of this study is to develop and test deep learning algorithms from diagnostic H&E images of RMS tumors which can aid in the diagnosis, mutation prediction and risk stratification for RMS patients. Methods: De-identified RMS patient samples were collected from tissue banking studies from Children’s Oncology Group (n=275), University Hospital Zurich (n=250) and Memorial Sloan Kettering (n=10). H&E stains on whole slides or TMAs were digitally scanned and used for analysis. Clinical information including clinical risk group, event-free survival, and genomic findings were used as available for training and testing. Convolutional neural networks (CNN) using EfficientNet were trained using K-fold cross validation to classify tumor histology, mutation probability, and risk stratification and tested against randomly selected samples or independent datasets when available. Results: The developed AI algorithm was able to classify tissue as ARMS (FP-RMS), ERMS (FN-RMS), stroma and necrosis with an average weighted intersect-over-union of 0.74 when compared to an expert pathologist annotation. A second deep learning algorithm developed specifically for distinguishing FP-RMS from FN-RMS displayed excellent sensitivity (FP-RMS=0.88) and specificity (FP-RMS=0.86) when tested against an independent RMS TMA dataset. Algorithms were also trained to predict mutations in MYOD1, RAS pathway genes, and TP53 and displayed good performance with ROC values of 0.96, 0.68, and 0.64, respectively. Lastly, we developed an algorithm to provide a Cox proportional hazard prediction based on H&E images. The resulting algorithm was capable of predicting EFS with similar accuracy as current clinical risk group assessment with improved ability to distinguish intermediate and high risk patients. Conclusions: Deep learning with convolutional neural networks provides pathologist-independent classification of RMS patients from simple H&E images. These AI algorithms can provide probabilities of prognostically relevant genetic alterations and survival which will ultimately contribute to better risk stratification of RMS patients. Citation Format: David Milewski, Hyun Jung, G. Thomas Brown, Yanling Liu, Jack Collins, Marc Ladanyi, Erin Rudzinski, Javed Khan. Predicting survival of rhabdomyosarcoma patients based on deep learning of H&E images [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 466.

Background The cancer genome is commonly altered with thousands of structural rearrangements including insertions, deletions, translocation, inversions, duplications, and copy number variations. Thus, structural variant (SV) characterization plays a paramount role in cancer target identification, oncology diagnostics, and personalized medicine. As part of the SEQC2 Consortium effort, the present study established and evaluated a consensus SV call set using a breast cancer reference cell line and matched normal control derived from the same donor, which were used in our companion benchmarking studies as reference samples. Results We systematically investigated somatic SVs in the reference cancer cell line by comparing to a matched normal cell line using multiple NGS platforms including Illumina short-read, 10X Genomics linked reads, PacBio long reads, Oxford Nanopore long reads, and high-throughput chromosome conformation capture (Hi-C). We established a consensus SV call set of a total of 1788 SVs including 717 deletions, 230 duplications, 551 insertions, 133 inversions, 146 translocations, and 11 breakends for the reference cancer cell line. To independently evaluate and cross-validate the accuracy of our consensus SV call set, we used orthogonal methods including PCR-based validation, Affymetrix arrays, Bionano optical mapping, and identification of fusion genes detected from RNA-seq. We evaluated the strengths and weaknesses of each NGS technology for SV determination, and our findings provide an actionable guide to improve cancer genome SV detection sensitivity and accuracy. Conclusions A high-confidence consensus SV call set was established for the reference cancer cell line. A large subset of the variants identified was validated by multiple orthogonal methods.

Abstract Background: Rhabdomyosarcoma (RMS) is an aggressive soft tissue tumor in children and young adults, accounting for 350-400 new cases annually in the US. Diagnosis of RMS is defined by the expression of genes related to skeletal muscle differentiation and can be further subclassified based on histological patterns (embryonal, ERMS; alveolar, ARMS; spindle/sclerosing, SSRMS). Genetic studies have found that the presence of a PAX fusion gene (FP-RMS), which is present in many ARMS tumors, correlates with poor outcome. Subsequent studies have identified additional genetic alterations (ex. TP53 or MYOD1 mutations) which also display distinct histological features and are associated with poor outcome. As a result, there is a growing need to identify these mutations to improve risk stratification. The goal of this study is to develop and test deep learning algorithms from diagnostic H&E images of RMS tumors which can aid in the diagnosis, mutation prediction and risk stratification for RMS patients. Methods: De-identified RMS patient samples were collected from tissue banking studies from Children’s Oncology Group (n=275), University Hospital Zurich (n=250) and Memorial Sloan Kettering (n=10). H&E stains on whole slides or TMAs were digitally scanned and used for analysis. Clinical information including clinical risk group, event-free survival, and genomic findings were used as available for training and testing. Convolutional neural networks (CNN) using EfficientNet were trained using K-fold cross validation to classify tumor histology, mutation probability, and risk stratification and tested against randomly selected samples or independent datasets when available. Results: The developed AI algorithm was able to classify tissue as ARMS (FP-RMS), ERMS (FN-RMS), stroma and necrosis with an average weighted intersect-over-union of 0.74 when compared to an expert pathologist annotation. A second deep learning algorithm developed specifically for distinguishing FP-RMS from FN-RMS displayed excellent sensitivity (FP-RMS=0.88) and specificity (FP-RMS=0.86) when tested against an independent RMS TMA dataset. Algorithms were also trained to predict mutations in MYOD1, RAS pathway genes, and TP53 and displayed good performance with ROC values of 0.96, 0.68, and 0.64, respectively. Lastly, we developed an algorithm to provide a Cox proportional hazard prediction based on H&E images. The resulting algorithm was capable of predicting EFS with similar accuracy as current clinical risk group assessment with improved ability to distinguish intermediate and high risk patients. Conclusions: Deep learning with convolutional neural networks provides pathologist-independent classification of RMS patients from simple H&E images. These AI algorithms can provide probabilities of prognostically relevant genetic alterations and survival which will ultimately contribute to better risk stratification of RMS patients. Citation Format: David Milewski, Hyun Jung, G. Thomas Brown, Yanling Liu, Jack Collins, Marc Ladanyi, Erin Rudzinski, Javed Khan. Predicting survival of rhabdomyosarcoma patients based on deep learning of H&E images [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 466.

A composite material of plasmonic nanoparticles embedded in a scaffold of nano-porous silicon offers unmatched capabilities for use as a SERS substrate. The marriage of these components presents an exclusive combination of tightly focused amplification of Localised Surface Plasmon (LSP) fields inside the material with an extremely high surface-to-volume ratio. This provides favourable conditions for a single molecule or extremely low concentration detection by SERS. In this work the advantage of the composite is demonstrated by SERS detection of Methylene Blue at a concentration as low as a few picomolars. We systematically investigate the plasmonic properties of the material by imaging its morphology, establishing its composition and the effect on the LSP resonance optical spectra.

SUMMARY:The Annotation, Visualization and Impact Analysis (AVIA) is a web application combining multiple features to annotate and visualize genomic variant data. Users can investigate functional significance of their genetic alterations across samples, genes and pathways. Version 3.0 of AVIA offers filtering options through interactive charts and by linking disease relevant data sources. Newly incorporated services include gene, variant and sample level reporting, literature and functional correlations among impacted genes, comparative analysis across samples and against data sources such as TCGA and ClinVar, and cohort building. Sample and data management is now feasible through the application, which allows greater flexibility with sharing, reannotating and organizing data. Most importantly, AVIA's utility stems from its convenience for allowing users to upload and explore results without any a priori knowledge or the need to install, update and maintain software or databases. Together, these enhancements strengthen AVIA as a comprehensive, user-driven variant analysis portal. AVAILABILITYAND IMPLEMENTATION:AVIA is accessible online at https://avia-abcc.ncifcrf.gov.

COVID-19 ranges from asymptomatic in 35% of cases to severe in 20% of patients. Differences in the type and degree of inflammation appear to determine the severity of the disease. Recent reports show an increase in circulating monocytic-myeloid-derived suppressor cells (M-MDSC) in severe COVID 19 that deplete arginine but are not associated with respiratory complications. Our data shows that differences in the type, function and transcriptome of granulocytic-MDSC (G-MDSC) may in part explain the severity COVID-19, in particular the association with pulmonary complications. Large infiltrates by Arginase 1(+) G-MDSC (Arg(+)G-MDSC), expressing NOX-1 and NOX-2 (important for production of reactive oxygen species) were found in the lungs of patients who died from COVID-19 complications. Increased circulating Arg(+)G-MDSC depleted arginine, which impaired T cell receptor and endothelial cell function. Transcriptomic signatures of G-MDSC from patients with different stages of COVID-19, revealed that asymptomatic patients had increased expression of pathways and genes associated with type I interferon (IFN), while patients with severe COVID-19 had increased expression of genes associated with arginase production, and granulocyte degranulation and function. These results suggest that asymptomatic patients develop a protective type I IFN response, while patients with severe COVID-19 have an increased inflammatory response that depletes arginine, impairs T cell and endothelial cell function, and causes extensive pulmonary damage. Therefore, inhibition of arginase-1 and/or replenishment of arginine may be important in preventing/treating severe COVID-19.

COVID-19 ranges from asymptomatic in 35% of cases to severe in 20% of patients. Differences in the type and degree of inflammation appear to determine the severity of the disease. Recent reports show an increase in circulating monocytic-myeloid-derived suppressor cells (M-MDSC) in severe COVID 19, that deplete arginine but are not associated with respiratory complications. Our data shows that differences in the type, function and transcriptome of Granulocytic-MDSC (G-MDSC) may in part explain the severity COVID-19, in particular the association with pulmonary complications. Large infiltrates by Arginase 1 + G-MDSC (Arg + G-MDSC), expressing NOX-1 and NOX-2 (important for production of reactive oxygen species) were found in the lungs of patients who died from COVID-19 complications. Increased circulating Arg + G-MDSC depleted arginine, which impaired T cell receptor and endothelial cell function. Transcriptomic signatures of G-MDSC from patients with different stages of COVID-19, revealed that asymptomatic patients had increased expression of pathways and genes associated with type I interferon (IFN), while patients with severe COVID-19 had increased expression of genes associated with arginase production, and granulocyte degranulation and function. These results suggest that asymptomatic patients develop a protective type I IFN response, while patients with severe COVID-19 have an increased inflammatory response that depletes arginine, impairs T cell and endothelial cell function, and causes extensive pulmonary damage. Therefore, inhibition of arginase-1 and/or replenishment of arginine may be important in preventing/treating severe COVID-19.

Coupling between nanoplasmonics and semiconducting materials can enhance and complement the efficiency of almost all semiconductor technologies. It has been demonstrated that such composites enhance the light coupling to nanowires, increase photocurrent in detectors, enable sub-gap detection, allow DNA detection, and produce large non-linearity. Nevertheless, the tailored fabrication using the conventional methods to produce such composites remains a formidable challenge. This work attempts to resolve that deficiency by deploying the immersion-plating method to spontaneously grown gold clusters inside nano-porous silicon (np-Si). This method allows the fabrication of thin films of np-Si with embedded gold nanoparticles (Au) and creates nanoplasmonic-semiconductor composites, np-Si/Au, with fractional volume between 0.02 and 0.13 of the metallic component. Optical scattering measurements reveal a distinctive, 200 nm broad, localized surface plasmon (LSP) resonance, centered around 700 nm. Linear and non-linear properties, and their time evolution are investigated by optically pumping the LSP resonance and probing the optical response with short wavelength infra-red (2.5 mu m) light. The ultrafast time-resolved study demonstrates unambiguously that the non-linear response is not only directly related to the LSP excitation, but strongly enhanced with respect to bare np-Si, while its strength can be tuned by varying the metallic component.