Qlucore Newsletter: Machine learning at your fingertips
One click machine learning since 2016
With Qlucore Omics Explorer (QOE) you can without programming skills, test and build machine learning classifiers. A fast route to multi-omics capable companion diagnostics (CDx) applications with Qlucore Insights.
QOE contains built in machine learning functionality, available since 2016, which enables users without any need for programming to build classifiers and predict outcomes. Classifiers are built using frameworks such as Boosted trees, Support Vector Machines, Random trees, and kNN. A cross validation scheme enables integrated validation in addition to using an external data set. When a classifier is created, it can be used to classify a sample or samples in a second data set.
Qlucore classification methods use so called supervised machine learning techniques. The user chooses the class annotation in the training data set. Based on this class annotation, the classifier is “trained” to find an optimal subset of variables and parameters. The machine learning functionality can be used on any supported type of data.
Classification is also useful when looking for new therapeutic targets and biomarkers, as it often delivers shortlists of variables that collectively can correctly predict the class of new samples with an estimated accuracy. For example, the class annotation can be response vs. nonresponse or case vs. control.
Qlucore has further enhanced and integrated machine learning functionality into its solutions for precision and companion diagnostics – Qlucore Insights and Qlucore Diagnostics.
- No programming required.
- Build classifiers using several different frameworks such as Random trees or SVM.
- Automatic training of the.
- Capabilities extend to Qlucore Insights.
Register for the upcoming webinar:
Qlucore Omics Explorer: Basic Training
November 29th, 2023
15:00 GMT (+1)