Comprehensive machine-learning-based analysis of microRNA–target interactions reveals variable transferability of interaction rules across species

Table 8 Data processing pipeline

Source	[18]	[21]	[20]	[22]	[19]
Datasets	h1	ce1, h2, m1	ca1	ce2	h3, m2
miRNA sequence	\(\checkmark\)	\(\checkmark\)	miRBASE	miRBASE	miRBASE
Target sequence	\(\checkmark\)	\(\checkmark\)	\(\checkmark\)	Wormbase	UCSC genome browser
Site region	Ensembl Biomart + Blast
Duplex structure	Vienna RNAduplex
Seed filter	Canonical and non-canonical seeds only

The table describes the set of actions required to transform the datasets into a uniform format to serve as input for further data analysis and machine learning experiments. The check-mark sign (\(\checkmark\)) represents a piece of information taken directly from the paper without additional calculations

ISSN: 1471-2105