Table 4.

A summary of the analogies between document-topic-word and a biological object in the relevant studies (see ““Document-word-topic” in biological data” section)

Reference	Words	Topics	Documents	Biological dataset
Rogers et al. (2005), Masada et al. (2009), Perina et al. (2010), Bicego et al. (2010a, b, 2012), Lee et al. (2014)	Genes	Functional groups	Samples	Expression microarray data
Masseroli et al. (2012), Pinoli et al. (2013, 2014), Youngs et al. (2014)	Ontological terms	Latent relationship	Proteins	Protein annotations
Chen et al. (2010, 2012a, b), La Rosa et al. (2015), Zhang et al. (2015)	K-mers of DNA sequences	Taxonomic category/components of the whole genome	DNA sequences	Genomic sequences
Caldas et al. (2009)	Gene sets	Biological process	Experiments	Gene expression dataset
Coelho et al. (2010)	Object classes	Fundamental patterns	Images	Fluorescence images
Konietzny et al. (2011)	A fixed-sized vocabulary of words based on the gene annotations	Functional modules of protein families	Genome annotations	A set of genome annotations
Bisgin et al. (2013)	Endpoint measurements	Diagnostic topics	Drugs	Expression of the HCS endpoints
Chen et al. (2011), Randhave and Sonkamble (2014)	Functional elements (NCBI taxonomic level indicators, indicator of gene orthologous groups and KEGG pathway indicators)	Functional groups	Samples	Genome set
Pan et al. (2010)	Local sequential features	Latent topic features	Protein sequences	Protein–protein interaction dataset
Castellani et al. (2010)	Shape descriptors	Brain surface geometric patterns	Images	Magnetic resonance images
Pratanwanich and Lio (2014)	Genes	Pathways	Gene expression profiles	Gene expression data
Dawson and Kendziorski (2012)	Clinical events, treatment protocols, and genomic information from multiple sources	The category of patients	Patients	Patient’s text constructed from clinical and multidimensional genomic analyses