Response representation shapes AlphaGenome enhancer-gene prediction in human astrocytes
- 1University of Washington
Abstract
Background: Sequence-to-function models predict regulatory perturbation responses, but their added value beyond established enhancer-gene predictors and across prediction tasks remains unclear. We evaluated whole-enhancer deletion and sparse variant responses in human astrocytes. Results: We analyzed 2,307 CRISPR-interference-tested relations, including 133 expression-decreasing hits. Adding deletion responses increased fully nested EGrf average precision from 0.5475 to 0.6137. The paired increment was 0.0662 (enhancer-cluster 95% interval 0.0144-0.1181). In a separate exploratory analysis, sparse Atlas responses were combined with seven-input logistic and random-forest comparators by equal-weight rank fusion. Atlas magnitude fusion exceeded absolute-deletion fusion by 0.0519 and 0.0637 in average precision. Enhancer-cluster 95% intervals were 0.0098-0.0957 and 0.0220-0.1094, respectively. Standalone Atlas magnitude reached average precision 0.5570, but its advantage over absolute deletion remained uncertain. Signed Atlas fusion performed worse than either comparator alone. Incremental value for within-enhancer target choice and in the external WTC11 MHC evaluation remained uncertain. The reported performance intervals use one-way clustering of fixed predictions; Atlas analyses were exploratory within the same cohort. Conclusions: In AstroREG, whole-enhancer deletion responses increased observed retrieval performance beyond a fully nested chromatin-based predictor. Sparse variant responses were more informative as magnitudes than as signed averages. Improved within-enhancer target choice and external transfer were not established.
Related
- AlphaGenome informed machine learning for pathogenicity prediction of germline mutations
- A biological-response compound representation allows chemical perturbation prediction across cell lines
- Scooby-2: fine-tuning AlphaGenome for single-cell genomic track prediction
- Characterisation of cis-regulatory elements and genome wide enhancer prediction in Heliconiini butterflies
- Tri-Modality Representation Learning for Molecular Property Prediction
BibTeX
@misc{huang2026response,
title = {{Response representation shapes AlphaGenome enhancer-gene prediction in human astrocytes}},
author = {Z. Huang and R. Huang and J. Han},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.14.744988},
url = {https://www.biorxiv.org/content/10.64898/2026.08.14.744988v2},
note = {CC BY},
}
RIS
TY - UNPB
TI - Response representation shapes AlphaGenome enhancer-gene prediction in human astrocytes
AU - Z. Huang
AU - R. Huang
AU - J. Han
PY - 2026
DA - 2026/08/14
PB - bioRxiv
DO - 10.64898/2026.08.14.744988
UR - https://www.biorxiv.org/content/10.64898/2026.08.14.744988v2
AB - Background: Sequence-to-function models predict regulatory perturbation responses, but their added value beyond established enhancer-gene predictors and across prediction tasks remains unclear. We evaluated whole-enhancer deletion and sparse variant responses in human astrocytes. Results: We analyzed 2,307 CRISPR-interference-tested relations, including 133 expression-decreasing hits. Adding deletion responses increased fully nested EGrf average precision from 0.5475 to 0.6137. The paired increment was 0.0662 (enhancer-cluster 95% interval 0.0144-0.1181). In a separate exploratory analysis, sparse Atlas responses were combined with seven-input logistic and random-forest comparators by equal-weight rank fusion. Atlas magnitude fusion exceeded absolute-deletion fusion by 0.0519 and 0.0637 in average precision. Enhancer-cluster 95% intervals were 0.0098-0.0957 and 0.0220-0.1094, respectively. Standalone Atlas magnitude reached average precision 0.5570, but its advantage over absolute deletion remained uncertain. Signed Atlas fusion performed worse than either comparator alone. Incremental value for within-enhancer target choice and in the external WTC11 MHC evaluation remained uncertain. The reported performance intervals use one-way clustering of fixed predictions; Atlas analyses were exploratory within the same cohort. Conclusions: In AstroREG, whole-enhancer deletion responses increased observed retrieval performance beyond a fully nested chromatin-based predictor. Sparse variant responses were more informative as magnitudes than as signed averages. Improved within-enhancer target choice and external transfer were not established.
ER -