✨PromoterAI

Overview

PromoterAI is an AI annotation model developed by the Illumina Artificial Intelligence Lab to predict impact of variants in promoter regions.

The model evaluates DNA sequence context around a promoter variant and estimates whether the variant is likely to increase, decrease, or have no effect on gene expression.

Promoter variants are estimated to account for approximately 6% of genetic factors underlying rare disease, highlighting the importance of analyzing non-coding regulatory regions alongside coding variants.

For more details, refer to:

Publication

Jaganathan, et al. Predicting expression-altering promoter mutations with deep learning. Science (2025). https://doi.org/10.1126/science.ads7373

Professional data source

This data source requires a Professional license. Contact [e-mail] to request access.

Parsing

TSV File

gene_id	chrom	pos	ref	alt	strand	gene	transcript_id	tss_pos	strand_adj_dist_tss	score_x	score_y	score
ENSG00000181404	chr9	73364	G	A	-1	WASHC1	ENST00000642633.1	73864	500	-0.0032	-0.0298	-0.0165
ENSG00000181404	chr9	73364	G	C	-1	WASHC1	ENST00000642633.1	73864	500	-0.0372	-0.052000000000000005	-0.0446
ENSG00000181404	chr9	73364	G	T	-1	WASHC1	ENST00000642633.1	73864	500	0.0202	-0.0315	-0.0056500000000000005
ENSG00000181404	chr9	73365	G	A	-1	WASHC1	ENST00000642633.1	73864	499	-0.015	-0.0643	-0.03965
ENSG00000181404	chr9	73365	G	C	-1	WASHC1	ENST00000642633.1	73864	499	0.0285	-0.0035	0.0125
ENSG00000181404	chr9	73365	G	T	-1	WASHC1	ENST00000642633.1	73864	499	-0.0095	-0.0352	-0.022350000000000002
ENSG00000181404	chr9	73366	G	A	-1	WASHC1	ENST00000642633.1	73864	498	0.0068	-0.0565	-0.02485
ENSG00000181404	chr9	73366	G	C	-1	WASHC1	ENST00000642633.1	73864	498	0.0596	0.0854	0.07250000000000001
ENSG00000181404	chr9	73366	G	T	-1	WASHC1	ENST00000642633.1	73864	498	-0.0086	-0.0009	-0.00475
ENSG00000181404	chr9	73367	C	A	-1	WASHC1	ENST00000642633.1	73864	497	0.0091	-0.0068	0.0011500000000000004
...

From the file, Illumina Connected Annotations extracts these columns:

• gene_id
• transcript_id
• strand
• strand_adj_dist_tss
• score

JSON output

"promoterAI": [
    {
      "strand": 1,
      "distanceFromTss": -292,
      "geneId": "ENSG00000274391",
      "transcriptId": "ENST00000618007.5",
      "score": 0.032
    }
]

Field	Type	Notes
strand	int	Strand location of the transcript
distanceFromTss	int	Number of nucleotides calculated from TSS of the corresponding transcript
geneId	string	Gene ID
transcriptId	string	Transcript ID (Ensembl)
score	decimal	Calculated PromoterAI score

Transcript and gene IDs from PromoterAI

Illumina Connected Annotations uses its own transcript and gene cache data when reporting transcript annotations. These IDs can differ from the IDs reported directly by PromoterAI, which is based on GENCODE Release 39. As a result, transcript IDs and gene IDs can differ between the source data and the final annotation output.

Interpreting scores

PromoterAI scores range from -1 to 1.

• A positive score suggests increased expression of the target gene.
• A negative score suggests decreased expression of the target gene.
• Scores between -0.05 and 0.05 suggest no meaningful effect on target gene expression.

Resources

• PromoterAI GitHub
• GENCODE Release 39
• Ensembl release 105