Table 3.

Genomic tools/algorithm based on deep learning architecture for gene expression regulation

Tools	DL model	Application	Input/Output	Website Code Source	References
DanQ	CNN + BLSTM	To predict DNA function directly from sequence data	.mat /.mat	https://github.com/uci-cbcl/DanQ	[152]
SPEID	CNN + LSTM	For enhancer–promoter interaction (EPI) prediction	.mat /.mat	https://github.com/ma-compbio/SPEID	[153]
EP2vec	NLP + GBRT	To predict enhancer–promoter interactions (EPIs)	CSV / CSV	https://github.com/wanwenzeng/ep2vec	[154]
D-GEX (deep learning for gene expression)	FNN	To understand the expression of target genes from the expression of landmark genes	.cel, txt, BAM / txt	https://github.com/uci-cbcl/D-GEX	[155]
DeepExpression	CNN	To predict gene expression using promoter sequences and enhancer–promoter interactions	.txt /.txt	https://github.com/wanwenzeng/DeepExpression	[156]
DeepGSR	CNN + ANN	To recognise various types of genomic signals and regions (GSRs) in genomic DNA (e.g. splice sites and stop codon)	FASTA /.txt	https://zenodo.org/record/1117159#.Xp4B4y2B1p8	[157]
SpliceAI	CNN	To identify splice function from pre-mRNA sequencing	VCF / VCF	https://github.com/Illumina/SpliceAI	[71]
SpliceRover	CNN	For splice site prediction	FASTA /.txt	N/A	[158]
Splice2Deep	CNN	For splice site prediction in Genomic DNA	FASTA /.txt	https://github.com/SomayahAlbaradei/Splice_Deep	[29]
DeepBind	CNN	To characterise DNA- and RNA-binding protein specificity	FASTA /.txt	https://github.com/MedChaabane/DeepBind-with-PyTorch	[111]
Gene2vec	NLP	To produce a representation of genes distribution and predict gene–gene interaction	.txt /.txt	https://github.com/jingcheng-du/Gene2vec	[130]
MPRA-DragoNN	CNN	To predict and analyse the regulatory DNA sequences and non-coding genetic variants	N/A	https://github.com/kundajelab/MPRA-DragoNN	[77]
BiRen	CNN + GRU + RNN	For enhancers predictions	BED, BigWig /CSV	https://github.com/wenjiegroup/BiRen	[159]
APARENT (APA REgression NeT)	CNN	To predict and engineer the human 3' UTR Alternative Polyadenylation (APA) and annotate pathogenetic variants	FASTA / CSV	https://github.com/johli/aparent	[72]
LaBranchoR (LSTM Branchpoint Retriever)	BLSTM	To predict the location of RNA splicing branchpoint	FASTA / FASTA	https://github.com/jpaggi/labranchor	[160]
COSSMO	CNN, BLSTM + ResNet	To predict the splice site sequencing and splice factors	TSV, CSV /CSV	http://cossmo.genes.toronto.edu/	[79]
Xpresso	CNN	To predict gene expression levels from genomic sequence	FASTA /.txt	https://github.com/vagarwal87/Xpresso	[73]
DeepLoc	CNN + BLSTM	To predict subcellular localisation of protein from sequencing data	FASTA/ prediction score	https://github.com/JJAlmagro/subcellular_localization	[161]
SPOT-RNA	CNN	To predict RNA Secondary Structure	FASTA /.bpseq,.ct, and.prob	https://github.com/jaswindersingh2/SPOT-RNA/	[162]
DeepCLIP	CNN + BLSTM	For predicting the effect of mutations on protein–RNA binding	FASTA /.txt	https://github.com/deepclip/deepclip	[163]
DECRES (DEep learning for identifying Cis-Regulatory ElementS)	MLP + CNN	To predict active enhancers and promoters across the human genome	FASTA /.txt	https://github.com/yifeng-li/DECRES	[74]
DeepChrome	CNN	For prediction of gene expression levels from histone modification data	Bam / TSV	https://github.com/QData/DeepChrome	[164]
DARTS	DNN + BHT	Deep learning augmented RNA-seq analysis of transcript splicing	.txt	https://github.com/Xinglab/DARTS