A machine learning approach for the identification of key markers involved in brain development from single-cell transcriptomic data

Table 1 Genes/features selected by disparate feature selection techniques

Feature selection techniques^a	Features/Genes (No.)
Filtered by low expression	8281
GSVA feature enrichment	1161
sRAP	837
SVM-RFE	38
RF-based Positive MDA	3339
T-test	60

^aFeature selection is based on five different methodologies based on machine learning algorithms (SVM and RF) and also that of traditional differentially expressed genes (sRAP), t-test based analysis (limma) and genes in deregulated pathways (GSVA)

ISSN: 1471-2164