Foundation models have made significant strides in understanding the genomic language of DNA sequences. However, previous models typically adopt the tokenization methods designed for natural language, which are unsuitable for DNA sequences due to their unique characteristics. In addition, the optimal approach to tokenize DNA remains largely under-explored, and may not be intuitively understood by humans even if discovered. To address these challenges, we introduce MxDNA, a novel framework where the model autonomously learns an effective DNA tokenization strategy through gradient decent. MxDNA employs a sparse Mixture of Convolution Experts coupled with a deformable convolution to model the tokenization process, with the discontinuous, overlapping, and ambiguous nature of meaningful genomic segments explicitly considered. On Nucleotide Transformer Benchmarks and Genomic Benchmarks, MxDNA demonstrates superior performance to existing methods with less pretraining data and time, highlighting its effectiveness. Finally, we show that MxDNA learns unique tokenization strategy distinct to those of previous methods and captures genomic functionalities at a token level during self-supervised pretraining. Our MxDNA aims to provide a new perspective on DNA tokenization, potentially offering broad applications in various domains and yielding profound insights.

The accurate identification of active sites in proteins is essential for the advancement of life sciences and pharmaceutical development, as these sites are of critical importance for enzyme activity and drug design. Recent advancements in protein language models (PLMs), trained on extensive datasets of amino acid sequences, have significantly improved our understanding of proteins. However, compared to the abundant protein sequence data, functional annotations, especially precise per-residue annotations, are scarce, which limits the performance of PLMs. On the other hand, textual descriptions of proteins, which could be annotated by human experts or a pretrained protein sequence-to-text model, provide meaningful context that could assist in the functional annotations, such as the localization of active sites. This motivates us to construct a ProTein-Attribute text Dataset (ProTAD), comprising over 570,000 pairs of protein sequences and multi-attribute textual descriptions.

Proteins are the main machinery of life and protein functions are largely determined by their 3D structures. The measurement of the pairwise proximity between amino acids of a protein, known as inter-residue contact map, well characterizes the structural information of a protein. Protein contact prediction (PCP) is an essential building block of many protein structure related applications. The prevalent approach to contact prediction is based on estimating the inter-residue contacts using hand-crafted coevolutionary features derived from multiple sequence alignments (MSAs). To mitigate the information loss caused by hand-crafted features, some recently proposed methods try to learn residue co-evolutions directly from MSAs.

Figure X: Accuracy and feature importance ranking (FIR) scores yielded by different methods on the Enhancer-Promoter dataset: GM12878 Cell line (200 bp). While DFS and RF yield only one result with all the 102 features, RFE and ours produce the results at different subset sizes for s = 15, 25, 35, 45, 55. Note that the results yielded RFE and ours for s = 35 above are not specified deliberately with the subset size to indicate that those have been reported in the main text.

However, associating small molecules with changes in cell state is challenging.

A.1 Access instructions OpenProteinSet is hosted by the Registry of Open Data on AWS (RODA) and can be accessed at the following link: registry.opendata.aws/openfold/. A.2 Documentation and intended uses We include a datasheet [1] in Section B. Detailed documentation on the precise structure and content of the dataset is provided on the dataset's landing page. A.3 Data format All OpenProteinSet files are in standard plaintext formats (A3M for MSAs, HHSearch format for template hits, and PDB for structure files) that can be read by a wide variety of bioinformatics software. A.5 License OpenProteinSet is made available under the CC BY 4.0 license. A copy of the license is provided with the dataset.

Multiple sequence alignments (MSAs) of proteins encode rich biological information and have been workhorses in bioinformatic methods for tasks like protein design and protein structure prediction for decades. Recent breakthroughs like AlphaFold2 that use transformers to attend directly over large quantities of raw MSAs have reaffirmed their importance. Generation of MSAs is highly computationally intensive, however, and no datasets comparable to those used to train AlphaFold2 have been made available to the research community, hindering progress in machine learning for proteins. To remedy this problem, we introduce OpenProteinSet, an open-source corpus of more than 16 million MSAs, associated structural homologs from the Protein Data Bank, and AlphaFold2 protein structure predictions. We have previously demonstrated the utility of OpenProteinSet by successfully retraining AlphaFold2 on it. We expect OpenProteinSet to be broadly useful as training and validation data for 1) diverse tasks focused on protein structure, function, and design and 2) large-scale multimodal machine learning research.

Phenotype imputation plays a crucial role in improving comprehensive and accurate medical evaluation, which in turn can optimize patient treatment and bolster the reliability of clinical research. Despite the adoption of various techniques, multi-modal biological data, which can provide crucial insights into a patient's overall health, is often overlooked. With multi-modal biological data, patient characterization can be enriched from two distinct views: the biological view and the phenotype view. However, the heterogeneity and imprecise nature of the multimodal data still pose challenges in developing an effective method to model from two views. In this paper, we propose a novel framework to incorporate multi-modal biological data via view decoupling.