Table 2.

Genomic tools/algorithm based on deep learning architecture for disease variants

Tools	DL model	Application	Input/Output	Website Code Source	References
DeepPVP (PhenomeNet Variant Predictor)	ANN	to identify the variants in both whole exome or whole genome sequence data	VCF / VCF	https://github.com/bio-ontology-research-group/phenomenet-vp	[61]
ExPecto	CNN	Accurately predict tissue-specific transcriptional effects of mutations/functional SNPs	VCF/ CSV	https://github.com/FunctionLab/ExPecto	[138]
PEDIA (Prioritisation of exome data by image analysis)	CNN	To prioritise variants and genes for diagnosis of patients with rare genetic disorders	VCF / CSV	https://github.com/PEDIA-Charite/PEDIA-workflow	[148]
DeepMILO (Deep learning for Modeling Insulator Loops)	CNN + RNN	to predict the impact of non-coding sequence variants on 3D chromatin structure	FASTA / TSV	https://github.com/khuranalab/DeepMILO	[119]
DeepWAS	CNN	To identify disease or trait-associated SNPs	TSV / TSV	https://github.com/cellmapslab/DeepWAS	[19]
PrimateAI	CNN	To classify the pathogenicity of missense mutations	CSV / CSV + txt	https://github.com/Illumina/PrimateAI	[27]
DeepGestalt	CNN	To Identifying facial phenotypes of genetic disorders	Image / txt	Is available through the Face2Gene application, http://face2gene.com	[149]
DeepMiRGene	RNN, LSTM	To predict miRNA precursor	FASTA / Cross-Validation (CV)-Splits file	https://github.com/eleventh83/deepMiRGene	[150]
Basset	CNN	To predict the causative SNP with sets of related variants	BED, FASTA/ VCF	https://github.com/davek44/Basset	[151]