SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2412.05430 · NeurIPS · 2024

DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA

Austin Wang, Aman Patel, Arpita Singhal, Anusri Pampari, Maya Kasowski, Anshul Kundaje, Nnnelementnnn Nnnenlmteennn, Peak Sequences

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 16 functions out of this paper's own repositories and ran 1 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
kundajelab/DART-Eval canonical 1 of 1
kundajelab/dart-eval canonical 0 of 15
FunctionStatusWhere it lives
make_region_df Ran kundajelab/DART-Eval/src/dnalm_bench/task_1_paired_control/dataset_generators/encode_ccre.py
pointer only (licence: NONE) · get_code("853d8aee53d39296")
dinucleotide_shuffle Not yet run kundajelab/dart-eval/src/dnalm_bench/utils.py
pointer only (licence: NONE) · get_code("44434c209290ce5b")
evaluate_finetuned_classifier Not yet run kundajelab/dart-eval/src/dnalm_bench/task_1_paired_control/finetune.py
pointer only (licence: NONE) · get_code("f8b8d077d64721d3")
generate_latex_table_per_cell_type Not yet run kundajelab/dart-eval/src/dnalm_bench/task_1_paired_control/generate_latex.py
pointer only (licence: NONE) · get_code("b59c3fbf640d6503")
get_accuracies Not yet run kundajelab/dart-eval/src/dnalm_bench/task_2_5_single/experiments/task_2_transcription_factor_binding/footprint_eval_likelihoods.py
pointer only (licence: NONE) · get_code("8d44605ff38fcc40")
get_confidence_interval Not yet run kundajelab/dart-eval/src/dnalm_bench/task_1_paired_control/generate_latex.py
pointer only (licence: NONE) · get_code("3f5f87a9d18060a3")
get_distances Not yet run kundajelab/dart-eval/src/dnalm_bench/task_2_5_single/experiments/task_2_transcription_factor_binding/footprint_eval_embeddings.py
pointer only (licence: NONE) · get_code("ba9e7c54f2991e6a")
get_likelihoods Not yet run kundajelab/dart-eval/src/dnalm_bench/task_2_5_single/experiments/task_2_transcription_factor_binding/footprint_eval_likelihoods.py
pointer only (licence: NONE) · get_code("129082c04db6596b")
get_pvals Not yet run kundajelab/dart-eval/src/dnalm_bench/task_2_5_single/experiments/task_2_transcription_factor_binding/footprint_eval_likelihoods.py
pointer only (licence: NONE) · get_code("8dcd29ed5e28d719")
load_embeddings Not yet run kundajelab/dart-eval/src/dnalm_bench/task_2_5_single/experiments/task_2_transcription_factor_binding/footprint_eval_embeddings.py
pointer only (licence: NONE) · get_code("b05bd25300b743b3")
load_embeddings_and_labels Not yet run kundajelab/dart-eval/src/dnalm_bench/embedding_clustering.py
pointer only (licence: NONE) · get_code("ebb6d12a7eaca88b")
load_embeddings_and_labels_subset Not yet run kundajelab/dart-eval/src/dnalm_bench/embedding_clustering.py
pointer only (licence: NONE) · get_code("11d6f70d2317495f")
one_hot_encode Not yet run kundajelab/dart-eval/src/dnalm_bench/utils.py
pointer only (licence: NONE) · get_code("a32416b3ff7b6896")
onehot_to_chars Not yet run kundajelab/dart-eval/src/dnalm_bench/utils.py
pointer only (licence: NONE) · get_code("75f2134f65415576")
relate_embeddings_to_motifs Not yet run kundajelab/dart-eval/src/dnalm_bench/task_2_5_single/experiments/task_2_transcription_factor_binding/footprint_eval_embeddings.py
pointer only (licence: NONE) · get_code("ae9202aca7e7c370")
underline_max_values Not yet run kundajelab/dart-eval/src/dnalm_bench/task_1_paired_control/generate_latex.py
pointer only (licence: NONE) · get_code("51187133335595b3")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recent advances in self-supervised models for natural language, vision, and protein sequences have catalyzed the development of genomic DNA language models (DNALMs). These models aim to learn generalizable representations of diverse DNA elements, potentially enabling various downstream genomic prediction, interpretation and design tasks. However, existing benchmarks do not adequately assess the capabilities of DNALMs on an important class of non-coding DNA elements critical for regulating gene activity. Here, we introduce DART-Eval, a suite of representative benchmarks focused on regulatory DNA to evaluate performance of DNALMs across zero-shot, probed, and fine-tuned settings against contemporary ab initio models as baselines. DART-Eval addresses biologically relevant tasks including sequence motif discovery, cell-type specific regulatory activity prediction, and counterfactual prediction of regulatory genetic variants. Our systematic evaluations reveal that current annotation-agnostic DNALMs exhibit inconsistent performance and do not offer compelling gains over alternative baseline models for most tasks, despite requiring significantly more computational resources. We discuss potentially promising modeling, data curation, and evaluation strategies for the next generation of DNALMs. Our benchmark datasets and evaluation framework are available at https://github.com/kundajelab/DART-Eval 38th Conference on Neural Information Processing Systems (NeurIPS 2024) Track on Datasets and Benchmarks.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2412.05430")
get_code_for_paper("2412.05430")
have("2412.05430")

Connect an agent — have() is free.