Figure 1.

Overview of bootstraping performed by FeaLect. A row and a column of the gray data matrix correspond to a feature and a case, accordingly. 1000 models are trained, each fitted to a random subset that contains <a onClick="popup('http://www.biomedcentral.com/1471-2164/14/S1/S14/mathml/M30','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2164/14/S1/S14/mathml/M30">View MathML</a> of cases using Lasso technique [1]. Without any assumption from a-priori knowledge, all features are included for training the models. Then the selected features are scored by computing an average vote (eq. 3) to select the most predictive ones.

Zare et al. BMC Genomics 2013 14(Suppl 1):S14   doi:10.1186/1471-2164-14-S1-S14