A machine learning classifier trained on cancer transcriptomes detects NF1 inactivation signal in glioblastoma

Way, Gregory P.; Allaway, Robert J.; Bouley, Stephanie J.; Fadul, Camilo E.; Sánchez, Yolanda; Greene, Casey S.

doi:10.1101/075382

Cited by 4 publications

(5 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A recent study used TCGA data to classify tumors for TP53 inactivation status and found that alterations in multiple genes phenocopy TP53 inactivation, indicating that TP53 mutation status alone is not necessary to infer inactivation of the pathway (Knijnenburg et al 2018). We used a machine learning algorithm to infer TP53 inactivation, as well as NF1 inactivation and Ras pathway activation, from transcriptomes of PDX tumors using classifiers previously trained on TCGA expression data (STAR methods) (Knijnenburg et al 2018; Way et al 2018; Way et al 2017). The TP53 (AUROC = 0.89) and NF1 (AUROC = 0.77) classifiers are both accurate compared to a shuffled gene expression baseline, but performance of the Ras classifier (AUROC = 0.58) was relatively poor (Figure 4A).…”

Section: Resultsmentioning

confidence: 99%

“…Further, we performed machine learning to classify tumors into TP53 and NF1 active or inactive and we suggest that these scores might be future biomarkers for drug response. These classifiers have been used to identify tumors that may respond to novel agents, including those that target tumors driven by NF1 loss (Way, 2017). Although these machine learning algorithms are not ready for the clinic, the next logical step is to use PDX models to test the predictive nature of classifiers so that in the future, interdisciplinary teams can identify tumors driven by TP53 and/or NF1 loss, evaluate, and compare multiple therapies in real time.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Genomic profiling of childhood tumor patient-derived xenograft models to enable rational clinical trial design

Rokita

Rathi

Cardenas

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Summary Accelerating cures for children with cancer remains an immediate challenge due to extensive oncogenic heterogeneity between and within histologies, distinct molecular mechanisms evolving between diagnosis and relapsed disease, and limited therapeutic options. To systematically prioritize and rationally test novel agents in preclinical murine models, researchers within the Pediatric Preclinical Testing Consortium are continuously developing patient-derived xenografts (PDXs) from high-risk childhood cancers, many refractory to current standard-of-care treatments. Here, we genomically characterize 261 PDX models from 29 unique pediatric cancer malignancies and demonstrate faithful recapitulation of histologies, subtypes, and refine our understanding of relapsed disease. Expression and mutational signatures are used to classify tumors for TP53 and NF1 inactivation, as well as impaired DNA repair. We anticipate that these data will serve as a resource for pediatric oncology drug development and guide rational clinical trial design for children with cancer. Highlights Multiplatform genomic analysis defines landscape of 261 pediatric cancer patient derived xenograft (PDX) models Pediatric patient derived xenografts faithfully recapitulate relapsed disease Inferred TP53 pathway inactivation correlates with pediatric cancer copy number burden Somatic mutational signatures predict impaired DNA repair across multiple histologies

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Genomic profiling of childhood tumor patient-derived xenograft models to enable rational clinical trial design

Rokita

Rathi

Cardenas

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Studying the influence of age and demographic characteristics on the development of NF1 clinical features has the potential to inform more personalized approaches to the identification of symptom "complexes" and ultimately the clinical management of children with NF1. As such, the application of modern computational approaches 43,44 to NF1 facilitated the development of exploratory predictive models with variable performance to identify patients with OPG, ADHD, and plexiform neurofibromas using demographic, clinical features, and EHR data recorded before the clinical manifestation of the feature. The variability in model performance demonstrated herein for diagnosis of OPG and ADHD is most reasonably explained by differences in disease presentation, diagnostic methodology, and differences in clinical expertise of the NF1 clinician.…”

Section: Discussionmentioning

confidence: 99%

Predictive Modeling for Clinical Features Associated With Neurofibromatosis Type 1

et al. 2021

View full text Add to dashboard Cite

Objective:Perform a longitudinal analysis of clinical features associated with Neurofibromatosis Type 1 (NF1) based on demographic and clinical characteristics, and to apply a machine learning strategy to determine feasibility of developing exploratory predictive models of optic pathway glioma (OPG) and attention-deficit/hyperactivity disorder (ADHD) in a pediatric NF1 cohort.Methods:Using NF1 as a model system, we perform retrospective data analyses utilizing a manually-curated NF1 clinical registry and electronic health record (EHR) information, and develop machine-learning models. Data for 798 individuals were available, with 578 comprising the pediatric cohort used for analysis.Results:Males and females were evenly represented in the cohort. White children were more likely to develop OPG (OR: 2.11, 95%CI: 1.11-4.00, p=0.02) relative to their non-white peers. Median age at diagnosis of OPG was 6.5 years (1.7-17.0), irrespective of sex. Males were more likely than females to have a diagnosis of ADHD (OR: 1.90, 95%CI: 1.33-2.70, p<0.001), and earlier diagnosis in males relative to females was observed. The gradient boosting classification model predicted diagnosis of ADHD with an AUROC of 0.74, and predicted diagnosis of OPG with an AUROC of 0.82.Conclusions:Using readily available clinical and EHR data, we successfully recapitulated several important and clinically-relevant patterns in NF1 semiology specifically based on demographic and clinical characteristics. Naïve machine learning techniques can be potentially used to develop and validate predictive phenotype complexes applicable to risk stratification and disease management in NF1.

show abstract

“…We accessed RNA-seq data from the UCSC Xena data browser on March 8th, 2016 and archived the data in Zenodo. 30 To facilitate training, we min-maxed scaled RNA-seq data to the range of 0 – 1. We used corresponding clinical data accessed from the Snaptron web server.…”

Section: Methodsmentioning

confidence: 99%

Extracting a Biologically Relevant Latent Space from Cancer Transcriptomes with Variational Autoencoders

Way

Greene

2017

Preprint

Self Cite

View full text Add to dashboard Cite

The Cancer Genome Atlas (TCGA) has profiled over 10,000 tumors across 33 different cancer-types for many genomic features, including gene expression levels. Gene expression measurements capture substantial information about the state of each tumor. Certain classes of deep neural network models are capable of learning a meaningful latent space. Such a latent space could be used to explore and generate hypothetical gene expression profiles under various types of molecular and genetic perturbation. For example, one might wish to use such a model to predict a tumor's response to specific therapies or to characterize complex gene expression activations existing in differential proportions in different tumors. Variational autoencoders (VAEs) are a deep neural network approach capable of generating meaningful latent spaces for image and text data. In this work, we sought to determine the extent to which a VAE can be trained to model cancer gene expression, and whether or not such a VAE would capture biologically-relevant features. In the following report, we introduce a VAE trained on TCGA pan-cancer RNA-seq data, identify specific patterns in the VAE encoded features, and discuss potential merits of the approach. We name our method "Tybalt" after an instigative, cat-like character who sets a cascading chain of events in motion in Shakespeare's "Romeo and Juliet". From a systems biology perspective, Tybalt could one day aid in cancer stratification or predict specific activated expression patterns that would result from genetic changes or treatment effects.

show abstract

A machine learning classifier trained on cancer transcriptomes detects NF1 inactivation signal in glioblastoma

Cited by 4 publications

References 57 publications

Genomic profiling of childhood tumor patient-derived xenograft models to enable rational clinical trial design

Genomic profiling of childhood tumor patient-derived xenograft models to enable rational clinical trial design

Predictive Modeling for Clinical Features Associated With Neurofibromatosis Type 1

Extracting a Biologically Relevant Latent Space from Cancer Transcriptomes with Variational Autoencoders

Contact Info

Product

Resources

About