SCRUM-Japan genesis virtual sequencing (VSQ) project: A novel algorithm combining deep learning (DL) with pathological diagnostics to enable the prediction of BRAF mutations and microsatellite instability (MSI) in advanced colorectal cancer (CRC).

2021 ◽  
Vol 39 (3_suppl) ◽  
pp. 112-112
Author(s):  
Satoshi Fujii ◽  
Daisuke Kotani ◽  
Masahiro Hattori ◽  
Nishihara Masato ◽  
Toshihide Shikanai ◽  
...  
Keyword(s):  
Prediction Models ◽  
High Accuracy ◽  
Cancer Genome ◽  
Morphological Features ◽  
Braf V600e ◽  
Next Generation ◽  
Braf Mutations ◽  
Genetic Abnormalities ◽  
Image Prediction ◽  
Novel Algorithm

112 Background: Numerous genetic and epigenetic abnormalities may lead to various morphologies of cancer. However, exactly which gene abnormality causes which morphology is unknown. The VSQ Project aims at investigating a novel algorithm by synergistically fusing DL technology and pathological diagnostics for the prediction of cancer genome abnormalities. This was achieved by elucidating the association between the morphological findings and genetic abnormalities, including BRAF V600E mutations and MSI status directly linked to the therapeutic strategies for advanced CRC patients (pts). Methods: Clinicopathological-genomic integrated DB derived from SCRUM-Japan GI-SCREEN, a nation-wide cancer genome screening project including CRC, were used. A total of 1,657 images of thin sections (one representative image per pt) cut from formalin-fixed and paraffin-embedded (FFPE) tissue specimens from primary or metastatic tumors with genetic abnormalities confirmed by next-generation sequencing (NGS) were investigated; 1,234 and 423 images (one per pt) were used for training and validation cohorts, respectively. First, we developed image-prediction models based on the morphological features precisely annotated by the single central pathologist, and then constructed the DL algorithms (gene-prediction models) that enabled the prediction of gene abnormalities by using images filtered by the image-prediction models. Results: We achieved high accuracy of AUC > 0.90 for 12 features among the 33 morphological features analyzed. Next, we created several DL algorithms that enabled the prediction of BRAF mutations and MSI. The prediction level reached a high accuracy of AUC = 0.955 for the BRAF mutations and AUC = 0.857 for MSI in the training cohort. We verified the AUCs in the validation cohort and achieved AUC = 0.831 and 0.883 for BRAF mutations and MSI, respectively. Conclusions: Our findings suggest that VSQ can appropriately predict BRAF mutation and MSI status in advanced CRC, potentially without performing NGS tests. VSQ may also enable prompt initiation of systemic treatments in CRC patients as well as establish an unprecedented next-generation pathology in the near future.

scholarly journals Frequency detection of BRAF V600E mutation in a cohort of pediatric langerhans cell histiocytosis patients by next-generation sequencing

2021 ◽  
Vol 16 (1) ◽  
Author(s):  
Shunqiao Feng ◽  
Lin Han ◽  
Mei Yue ◽  
Dixiao Zhong ◽  
Jing Cao ◽  
...  
Keyword(s):  
Next Generation Sequencing ◽  
Langerhans Cell Histiocytosis ◽  
Mutation Screening ◽  
Langerhans Cell ◽  
Braf V600e ◽  
P Value ◽  
Type I ◽  
Next Generation ◽  
Mutation Status ◽  
Generation Sequencing

Abstract Background Langerhans cell histiocytosis (LCH) is a rare neoplastic disease that occurs in both children and adults, and BRAF V600E is detected in up to 64% of the patients. Several studies have discussed the associations between BRAF V600E mutation and clinicopathological manifestations, but no clear conclusions have been drawn regarding the clinical significance of the mutation in pediatric patients. Results We retrieved the clinical information for 148 pediatric LCH patients and investigated the BRAF V600E mutation using next-generation sequencing alone or with droplet digital PCR. The overall positive rate of BRAF V600E was 60/148 (41%). The type of sample (peripheral blood and formalin-fixed paraffin-embedded tissue) used for testing was significantly associated with the BRAF V600E mutation status (p-value = 0.000 and 0.000). The risk of recurrence declined in patients who received targeted therapy (p-value = 0.006; hazard ratio 0.164, 95%CI: 0.046 to 0.583). However, no correlation was found between the BRAF V600E status and gender, age, stage, specific organ affected, TP53 mutation status, masses close to the lesion or recurrence. Conclusions This is the largest pediatric LCH study conducted with a Chinese population to date. BRAF V600E in LCH may occur less in East Asian populations than in other ethnic groups, regardless of age. Biopsy tissue is a more sensitive sample for BRAF mutation screening because not all of circulating DNA is tumoral. Approaches with low limit of detection or high sensitivity are recommended for mutation screening to avoid type I and II errors.

scholarly journals Financial Information Asymmetry: Using Deep Learning Algorithms to Predict Financial Distress

Symmetry ◽  
2021 ◽  
Vol 13 (3) ◽  
pp. 443
Author(s):  
Chyan-long Jan
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Information Asymmetry ◽  
Error Rate ◽  
Financial Distress ◽  
Prediction Models ◽  
High Accuracy ◽  
Financial Information ◽  
Financial Distress Prediction ◽  
Distress Prediction

Because of the financial information asymmetry, the stakeholders usually do not know a company’s real financial condition until financial distress occurs. Financial distress not only influences a company’s operational sustainability and damages the rights and interests of its stakeholders, it may also harm the national economy and society; hence, it is very important to build high-accuracy financial distress prediction models. The purpose of this study is to build high-accuracy and effective financial distress prediction models by two representative deep learning algorithms: Deep neural networks (DNN) and convolutional neural networks (CNN). In addition, important variables are selected by the chi-squared automatic interaction detector (CHAID). In this study, the data of Taiwan’s listed and OTC sample companies are taken from the Taiwan Economic Journal (TEJ) database during the period from 2000 to 2019, including 86 companies in financial distress and 258 not in financial distress, for a total of 344 companies. According to the empirical results, with the important variables selected by CHAID and modeling by CNN, the CHAID-CNN model has the highest financial distress prediction accuracy rate of 94.23%, and the lowest type I error rate and type II error rate, which are 0.96% and 4.81%, respectively.

scholarly journals Validation of a Next-Generation Sequencing Assay Targeting RNA for the Multiplexed Detection of Fusion Transcripts and Oncogenic Isoforms

2019 ◽  
Vol 144 (1) ◽  
pp. 90-98 ◽  
Author(s):  
Robyn T. Sussman ◽  
Amanda R. Oran ◽  
Carmela Paolillo ◽  
David Lieberman ◽  
Jennifer J. D. Morrissette ◽  
...  
Keyword(s):  
Next Generation Sequencing ◽  
Solid Tumors ◽  
Gold Standard ◽  
University Hospital ◽  
Fusion Partner ◽  
Next Generation ◽  
Fixed Tissue ◽  
Fusion Transcripts ◽  
Genetic Abnormalities ◽  
Generation Sequencing

Context.— Next-generation sequencing is a high-throughput method for detecting genetic abnormalities and providing prognostic and therapeutic information for patients with cancer. Oncogenic fusion transcripts are among the various classifications of genetic abnormalities present in tumors and are typically detected clinically with fluorescence in situ hybridization (FISH). However, FISH probes only exist for a limited number of targets, do not provide any information about fusion partners, cannot be multiplex, and have been shown to be limited in specificity for common targets such as ALK. Objective.— To validate an anchored multiplex polymerase chain reaction–based panel for the detection of fusion transcripts in a university hospital–based clinical molecular diagnostics laboratory. Design.— We used 109 unique clinical specimens to validate a custom panel targeting 104 exon boundaries from 17 genes involved in fusions in solid tumors. The panel can accept as little as 100 ng of total nucleic acid from PreservCyt-fixed tissue, and formalin-fixed, paraffin-embedded specimens with as little as 10% tumor nuclei. Results.— Using FISH as the gold standard, this assay has a sensitivity of 88.46% and a specificity of 95.83% for the detection of fusion transcripts involving ALK, RET, and ROS1 in lung adenocarcinomas. Using a validated next-generation sequencing assay as the orthogonal gold standard for the detection of EGFR variant III (EGFRvIII) in glioblastomas, the assay is 92.31% sensitive and 100% specific. Conclusions.— This multiplexed assay is tumor and fusion partner agnostic and will provide clinical utility in therapy selection for patients with solid tumors.

scholarly journals High prevalence of somatic MAP2K1 mutations in BRAF V600E–negative Langerhans cell histiocytosis

Blood ◽  
2014 ◽  
Vol 124 (10) ◽  
pp. 1655-1658 ◽  
Author(s):  
Noah A. Brown ◽  
Larissa V. Furtado ◽  
Bryan L. Betz ◽  
Mark J. Kiel ◽  
Helmut C. Weigelin ◽  
...  
Keyword(s):  
Targeted Therapy ◽  
Genome Sequencing ◽  
High Frequency ◽  
Langerhans Cell Histiocytosis ◽  
Langerhans Cell ◽  
Braf V600e ◽  
Braf Mutations ◽  
Key Points ◽  
High Prevalence

Key Points Targeted genome sequencing reveals high-frequency somatic MAP2K1 mutations in Langerhans cell histiocytosis. MAP2K1 mutations are mutually exclusive with BRAF mutations and may have implications for the use of BRAF and MEK targeted therapy.

scholarly journals Cutaneous Side Effects of BRAF Inhibitors in Advanced Melanoma: Review of the Literature

2016 ◽  
Vol 2016 ◽  
pp. 1-6 ◽  
Author(s):  
Bilgen Gençler ◽  
Müzeyyen Gönül
Keyword(s):  
Side Effects ◽  
Metastatic Melanoma ◽  
Mapk Pathway ◽  
Case Reports ◽  
Mitogen Activated Protein Kinase ◽  
Braf V600e ◽  
Braf Inhibitors ◽  
Braf Mutations ◽  
Cutaneous Side Effects ◽  
Downstream Signaling

The incidence of melanoma has recently been increasing. BRAF mutations have been found in 40–60% of melanomas. The increased activity of BRAF V600E leads to the activation of downstream signaling through the mitogen-activated protein kinase (MAPK) pathway, which plays a key role as a regulator of cell growth, differentiation, and survival. The use of BRAF inhibitors in metastatic melanoma with BRAF mutation ensures clinical improvement of the disease. Vemurafenib and dabrafenib are two selective BRAF inhibitors approved by the Food and Drug Administration (FDA). Both drugs are well tolerated and successfully used in clinical practice. However, some adverse reactions have been reported in patients in the course of treatment. Cutaneous side effects are the most common adverse events among them with a broad spectrum. Both the case reports and several original clinical trials reported cutaneous reactions during the treatment with BRAF inhibitors. In this review, the common cutaneous side effects of BRAF inhibitors in the treatment of metastatic melanoma with BRAF V600E mutation were reviewed.

Rapid Somatic Mutation Testing in Colorectal Cancer by Use of a Fully Automated System and Single-Use Cartridge: A Comparison with Next-Generation Sequencing

2018 ◽  
Vol 3 (2) ◽  
pp. 178-184 ◽  
Author(s):  
M Rabie Al-Turkmani ◽  
Kelley N Godwin ◽  
Jason D Peterson ◽  
Gregory J Tsongalis
Keyword(s):  
Colorectal Cancer ◽  
Next Generation Sequencing ◽  
Ffpe Tissue ◽  
Braf V600e ◽  
Automated System ◽  
Next Generation ◽  
Tissue Samples ◽  
Tissue Sections ◽  
Single Use ◽  
Generation Sequencing

AbstractBackgroundMolecular tests have been increasingly used in the management of various cancers as more targeted therapies are becoming available as treatment options. The Idylla™ system is a fully integrated, cartridge-based platform that provides automated sample processing (deparaffinization, tissue digestion, and DNA extraction) and real-time PCR-based mutation detection with all reagents included in a single-use cartridge. This retrospective study aimed at evaluating both the Idylla KRAS and NRAS-BRAF-EGFR492 Mutation Assay cartridges (research use only) against next-generation sequencing (NGS) by using colorectal cancer (CRC) tissue samples.MethodsForty-four archived formalin-fixed paraffin-embedded (FFPE) CRC tissue samples previously analyzed by targeted NGS were tested on the Idylla system. Among these samples, 17 had a mutation in KRAS proto-oncogene, GTPase (KRAS), 5 in NRAS proto-oncogene, GTPase (NRAS), and 12 in B-Raf proto-oncogene, serine/threonine kinase (BRAF) as determined using the Ion AmpliSeq 50-gene Cancer Hotspot Panel v2. The remaining 10 samples were wild-type for KRAS, NRAS, and BRAF. Two 10-μm FFPE tissue sections were used for each Idylla run, 1 for the KRAS cartridge, and 1 for the NRAS-BRAF-EGFR492 cartridge. All cases met the Idylla minimum tumor content requirement for KRAS, NRAS, and BRAF (≥10%). Assay reproducibility was evaluated by testing commercial controls derived from human cell lines, which had an allelic frequency of 50% and were run in triplicate.ResultsThe Idylla system successfully detected all mutations previously identified by NGS in KRAS (G12C, G12D, G12V, G13D, Q61K, Q61R, A146T), NRAS (G12V, G13R, Q61H), and BRAF (V600E). Compared with NGS, Idylla had a sensitivity of 100%. Analysis of the mutated commercial controls demonstrated agreement with the expected result for all samples and 100% reproducibility. The Idylla system produced results quickly with a turnaround time of approximately 2 h.ConclusionThe Idylla system offers reliable and sensitive testing of clinically actionable mutations in KRAS, NRAS, and BRAF directly from FFPE tissue sections.

scholarly journals A Next-Generation Sequencing Strategy for Evaluating the Most Common Genetic Abnormalities in Multiple Myeloma

2017 ◽  
Vol 19 (1) ◽  
pp. 99-106 ◽  
Author(s):  
Cristina Jiménez ◽  
María Jara-Acevedo ◽  
Luis A. Corchete ◽  
David Castillo ◽  
Gonzalo R. Ordóñez ◽  
...  
Keyword(s):  
Multiple Myeloma ◽  
Next Generation Sequencing ◽  
Next Generation ◽  
Genetic Abnormalities ◽  
Sequencing Strategy ◽  
Generation Sequencing

scholarly journals Learners Demographics Classification on MOOCs During the COVID-19: Author Profiling via Deep Learning Based on Semantic and Syntactic Representations

2021 ◽  
Vol 6 ◽  
Author(s):  
Tahani Aljohani ◽  
Alexandra I. Cristea
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Learning ◽  
Prediction Models ◽  
Short Term Memory ◽  
Methodological Approach ◽  
High Accuracy ◽  
Directional Model ◽  
Textual Representations ◽  
The One

Massive Open Online Courses (MOOCs) have become universal learning resources, and the COVID-19 pandemic is rendering these platforms even more necessary. In this paper, we seek to improve Learner Profiling (LP), i.e. estimating the demographic characteristics of learners in MOOC platforms. We have focused on examining models which show promise elsewhere, but were never examined in the LP area (deep learning models) based on effective textual representations. As LP characteristics, we predict here the employment status of learners. We compare sequential and parallel ensemble deep learning architectures based on Convolutional Neural Networks and Recurrent Neural Networks, obtaining an average high accuracy of 96.3% for our best method. Next, we predict the gender of learners based on syntactic knowledge from the text. We compare different tree-structured Long-Short-Term Memory models (as state-of-the-art candidates) and provide our novel version of a Bi-directional composition function for existing architectures. In addition, we evaluate 18 different combinations of word-level encoding and sentence-level encoding functions. Based on these results, we show that our Bi-directional model outperforms all other models and the highest accuracy result among our models is the one based on the combination of FeedForward Neural Network and the Stack-augmented Parser-Interpreter Neural Network (82.60% prediction accuracy). We argue that our prediction models recommended for both demographics characteristics examined in this study can achieve high accuracy. This is additionally also the first time a sound methodological approach toward improving accuracy for learner demographics classification on MOOCs was proposed.

scholarly journals NetGenes: A database of essential genes predicted using features from interaction networks

2020 ◽  
Author(s):  
Vimaladhasan Senthamizhan ◽  
Balaraman Ravindran ◽  
Karthik Raman
Keyword(s):  
Prediction Models ◽  
Essential Gene ◽  
Gene Prediction ◽  
High Accuracy ◽  
Essential Genes ◽  
Interaction Networks ◽  
Functional Association ◽  
String Database ◽  
Feature Vectors ◽  
Link Type

AbstractEssential gene prediction models built so far are heavily reliant on sequence-based features and the scope of network-based features has been narrow. Previous work from our group demonstrated the importance of using network-based features for predicting essential genes with high accuracy. Here, we applied our approach for the prediction of essential genes to organisms from the STRING database and hosted the results in a standalone website. Our database, NetGenes, contains essential gene predictions for 2700+ bacteria predicted using features derived from STRING protein-protein functional association networks. Housing a total of 3.5M+ genes, NetGenes offers various features like essentiality scores, annotations and feature vectors for each gene. NetGenes is available at https://rbc-dsai.iitm.github.io/NetGenes/

Sign in / Sign up

Export Citation Format

Share Document