Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: CRISPRpred(SEQ): a sequence-based method for sgRNA on target activity prediction using traditional machine learning

Fig. 1

Training pipelines, the steps of building the final prediction model. aThe pipeline for experimental setup A. We only extracted position-independent and position-specific features. The steps of splitting the dataset and selecting features are described in “Results” section. We used the default parameters while training SVM. bThe pipeline for experimental setup B. The steps of extracting features and splitting dataset is same as experimental setup A. But, in feature selection step we used extremely randomized trees (the feature selection criteria are described in “Results” section). We performed hyperparameter tuning on SVM and retrained SVM with the best hyperparameters. cThe pipeline for experimental setup C. It is exactly same as the pipeline for experimental setup B except we considered the feature type n-Gapped Di-nucleotide in feature extraction step

Back to article page