BEND: Benchmarking DNA Language Models on biologically meaningful tasks
This provides a standardized benchmark for researchers in genomics and bioinformatics to evaluate DNA language models, though it is incremental as it focuses on assessment rather than new model development.
The authors tackled the lack of standardized evaluation for DNA language models by introducing BEND, a benchmark with biologically meaningful tasks on the human genome, finding that current models approach expert methods on some tasks but have limited long-range feature capture.
The genome sequence contains the blueprint for governing cellular processes. While the availability of genomes has vastly increased over the last decades, experimental annotation of the various functional, non-coding and regulatory elements encoded in the DNA sequence remains both expensive and challenging. This has sparked interest in unsupervised language modeling of genomic DNA, a paradigm that has seen great success for protein sequence data. Although various DNA language models have been proposed, evaluation tasks often differ between individual works, and might not fully recapitulate the fundamental challenges of genome annotation, including the length, scale and sparsity of the data. In this study, we introduce BEND, a Benchmark for DNA language models, featuring a collection of realistic and biologically meaningful downstream tasks defined on the human genome. We find that embeddings from current DNA LMs can approach performance of expert methods on some tasks, but only capture limited information about long-range features. BEND is available at https://github.com/frederikkemarin/BEND.