MLSeq-class: 'MLSeq' object
In dncR/MLSeq: Machine Learning Interface for RNA-Seq Data

Description Details Slots Note Author(s) See Also

For classification, this is the main class for the MLSeq package. It contains all the information including trained model, selected genes, cross-validation results, etc.

Objects can be created by calls of the form new("MLSeq", ...). This type of objects is created as a result of classify function of MLSeq package. It is then used in predict or predictClassify function for predicting the class labels of new samples.

inputObject:: stores the data in DESeqDataSet object.
modelInfo:: stores all the information about classification model. The object is from subclass MLSeqModelInfo. See MLSeqModelInfo-class for details.
metaData:: metadata for MLSeq object. The object is from subclass MLSeqMetaData. See MLSeqMetaData-class for details.

An MLSeq class stores the results of classify function and offers further slots that are populated during the analysis. The slot inputObject stores the raw and transformed data throughout the classification. The slot modelInfo stores all the information about classification model. These results may contain the classification table and performance measures such as accuracy rate, sensitivity, specifity, positive and negative predictive values, etc. It also contains information on classification method, normalization and transformation used in the classification model. Lastly, the slot metaData stores the information about modified or updated slots in MLSeq object.

Dincer Goksuluk, Gokmen Zararsiz, Selcuk Korkmaz, Vahap Eldem, Ahmet Ozturk and Ahmet Ergun Karaagaoglu

MLSeqModelInfo-class, MLSeqMetaData-class

dncR/MLSeq documentation built on May 17, 2020, 6:45 p.m.