Abstract 5820
Background
DNA sequencing to identify variants is becoming increasingly valuable in clinical settings; including matching patients to approved targeted therapies, immunotherapies, and/or clinical trials. However, accurate calling of genetic variants from sequencing still remains challenging. With little corroboration between the different tools available, patients are at risk of being treated with therapies that are unsuitable for their cancer.
Methods
Here we present a novel machine learning based method for the accurate identification of somatic variants in cancer patient tumour samples, with a neural network architecture from encoded raw sequencing read information of tumour/normal sample pairings into an image, enabling it to classify whether a variant is germline, somatic, or sequencing error. The model was trained and tested on in-silico spike-in data using bam-surgeon, and then validated on a multi-cancer and multi-center dataset and benchmarked against industry standard variant callers.
Results
The approach, called somaticNET, outperforms existing industry standard tools in sensitivity and specificity, achieving an AUROC of ∼1.00 on the bam-surgeon dataset and an AUROC of ∼0.99 on the multi-cancer multicenter dataset. The model also works faster than other variant callers, in minutes compared to hours.
Conclusions
Using the power of machine learning for accurate somatic variant calling can improve patient matching to approved therapies and clinical trials, thus ensuring patients are given the right therapy at the right time to treat their cancer.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Cambridge Cancer Genomics.
Disclosure
G. Dubourg-Felonneau: Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. D. Rebergen: Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. C. Parsons: Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. H. Thompson: Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. J.W. Cassidy: Leadership role, Shareholder / Stockholder / Stock options, Full / Part-time employment, Officer / Board of Directors: Cambridge Cancer Genomics. N. Patel: Leadership role, Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics. H.W. Clifford: Leadership role, Shareholder / Stockholder / Stock options, Full / Part-time employment: Cambridge Cancer Genomics.
Resources from the same session
5822 - Greek nursing students experience facing death in clinical practice
Presenter: Maria Dimoula
Session: Poster Display session 3
Resources:
Abstract
2866 - HOPEVOL: Hospice care appropriate to the wishes and needs of patients in the palliative terminal phase.
Presenter: Merel van Klinken
Session: Poster Display session 3
Resources:
Abstract
829 - Mindfulness-based stress reduction in early palliative care for advanced cancer patients : an italian single-centre study. MINDEEP
Presenter: Emilia Gianotti
Session: Poster Display session 3
Resources:
Abstract
2702 - Optimising Inpatient Oncology Care
Presenter: Lisa Judge
Session: Poster Display session 3
Resources:
Abstract
1527 - Analysis on the Implementation Results of Family Sickbed for Oncology Patients in Dongshi Township Health Centers from 2015 to 2017
Presenter: Yayu Huang
Session: Poster Display session 3
Resources:
Abstract
2054 - Exploring needs for palliative care and quality of life for oncology patients with advanced disease who undergo radiotherapy
Presenter: Foteini Antonopoulou
Session: Poster Display session 3
Resources:
Abstract
5605 - Cytotoxic contamination in cancer care settings – Risks and safety awareness among cancer nurses
Presenter: Sandra Lundman Vikberg
Session: Poster Display session 3
Resources:
Abstract
5769 - Understanding Chemotherapy - group education sessions prior to commencing chemotherapy
Presenter: Aileen McHale
Session: Poster Display session 3
Resources:
Abstract
2620 - Estimation of HPQ-based absenteeism and presenteeism in cancer patients via ResearchKit
Presenter: Shunsuke Kondo
Session: Poster Display session 3
Resources:
Abstract
4705 - Identifying falls-related variables and risk factors in hospitalised cancer patients
Presenter: Maria Montserrat Martí Dillet
Session: Poster Display session 3
Resources:
Abstract