Skip to main content
AlphaGenome splice site usage score
License: AlphaGenome uses Apache-2.0 for code and Custom (AlphaGenome Terms of Use) for model weights and has restrictions around commercial use and may require explicit attribution when utilized. Model weights are gated and require accepting the provider’s terms and authenticating with a HuggingFace token. Please refer to the code license and model weights license for full terms.

This constraint is open source. Any third-party models, product names, or trademarks referenced are the property of their respective owners, and Proto is not affiliated with them.


Go to Tool Page
proto-bio/proto-language/proto_language/constraint/rna_splicing/alphagenome_splice_site_usage.py
View source
@article{avsec2026alphagenome,
  title={Advancing regulatory variant effect prediction with AlphaGenome},
  author={Avsec, {\v{Z}}iga and Latysheva, Natasha and Cheng, Jun and Novati, Guido and Taylor, Kyle R and Ward, Tom and Bycroft, Clare and Nicolaisen, Lauren and Arvaniti, Eirini and Pan, Joshua and Thomas, Raina and Dutordoir, Vincent and Perino, Matteo and De, Soham and Karollus, Alexander and Gayoso, Adam and Sargeant, Toby and Mottram, Anne and Wong, Lai Hong and Drot{\'a}r, Pavol and Kosiorek, Adam and Senior, Andrew and Tanburn, Richard and Applebaum, Taylor and Basu, Souradeep and Hassabis, Demis and Kohli, Pushmeet},
  journal={Nature},
  year={2026},
  volume={649},
  number={8099},
  pages={1206--1218},
  doi={10.1038/s41586-025-10014-0}
}
Copy citation
Score AlphaGenome SSU at selected positions in a three-part target. Each input tuple contains three DNA sequences (left_flank, intron_core, right_flank) which are concatenated into a target, wrapped with cassette contexts, and integrated into a genomic context for AlphaGenome prediction.

API Reference

ConfigAlphaGenomeSpliceSiteUsageConfig Source
Configuration for AlphaGenome splice-site-usage scoring.Takes three segments (left_flank, intron_core, right_flank), concatenates them into a target sequence, wraps with cassette contexts, and integrates into a genomic context for AlphaGenome prediction. Splice positions are specified relative to the concatenated target sequence.
genomic_context
string
required
Genomic context sequence for cassette integration (e.g., AAVS1 safe harbor locus).
cassette_left_context
string
required
Left flanking context for the cassette (plasmid/gene sequence 5’ of the target).
cassette_right_context
string
required
Right flanking context for the cassette (plasmid/gene sequence 3’ of the target).
ontology_terms
List[string]
required
AlphaGenome ontology term(s) to score.
splice_pos
List[integer]
required
0-indexed position(s) in the concatenated target to evaluate.
direction
enum
default:"max"
‘max’ returns 1-mean(SSU); ‘min’ returns mean(SSU).Options: max, min
strand
enum
default:"positive"
Track strand subset to aggregate over.Options: positive, negative, all
model_version
string
default:"all_folds"
AlphaGenome model version.
organism
enum
default:"human"
Organism for AlphaGenome prediction.Options: human, mouse
device
string
default:"cuda"
Device for AlphaGenome prediction.
prediction_timeout
integer
default:"3600"
Timeout (seconds) for each AlphaGenome prediction call.
ReturnsConstraintOutput
One result per input. score is in [0.0, 1.0] (interpretation depends on direction). metadata carries the selected tracks plus alphagenome_splice_site_usage_raw and _score.

Usage

python
from proto_language.core import Constraint
from proto_language.constraint import alphagenome_splice_site_usage, AlphaGenomeSpliceSiteUsageConfig

constraint = Constraint(
    inputs=[segment],
    function=alphagenome_splice_site_usage,
    function_config=AlphaGenomeSpliceSiteUsageConfig(
        # Configure parameters here
    ),
)

scores = constraint.evaluate()

Metadata

PropertyValue
Keyalphagenome-splice-site-usage
Functionalphagenome_splice_site_usage
Categoryrna_splicing
Modediscrete
Uses GPUTrue
Supported Typesdna