Monoclonal antibodies are becoming increasingly important therapeutic agents for the treatment of cancers, infectious diseases, and autoimmune disorders. However, laboratory-based methods of developing therapeutic monoclonal antibodies (e.g., immunized mice, hybridomas, and phage display) are time-consuming and are often unable to target a specific antigen epitope or reach (sub)nanomolar levels of affinity. To this end, we developed Optimal Method for Antibody Variable region Engineering (OptMAVEn) for de novo design of humanized monoclonal antibody variable regions targeting a specific antigen epitope. In this work, we introduce OptMAVEn-2.0, which improves upon OptMAVEn by (1) reducing computational resource requirements without compromising design quality; (2) clustering the designs to better identify high-affinity antibodies; and (3) eliminating intra-antibody steric clashes using an updated set of clashing parts from the Modular Antibody Parts (MAPs) database. Benchmarking on a set of 10 antigens revealed that OptMAVEn-2.0 uses an average of 74% less CPU time and 84% less disk storage relative to OptMAVEn. Testing on 54 additional antigens revealed that computational resource requirements of OptMAVEn-2.0 scale only sub-linearly with respect to antigen size. OptMAVEn-2.0 was used to design and rank variable antibody fragments targeting five epitopes of Zika envelope protein and three of hen egg white lysozyme. Among the top five ranked designs for each epitope, recovery of native residue identities is typically 45–65%. MD simulations of two designs targeting Zika suggest that at least one would bind with high affinity. OptMAVEn-2.0 can be downloaded from our GitHub repository and webpage as (links in Summary and Discussion section).

RNA molecules perform a diversity of essential functions for which their linear sequences must fold into higher-order structures. Techniques including crystallography and cryogenic electron microscopy have revealed 3D structures of ribosomal, transfer, and other well-structured RNAs; while chemical probing with sequencing facilitates secondary structure modeling of any RNAs of interest, even within cells. Ongoing efforts continue increasing the accuracy, resolution, and ability to distinguish coexisting alternative structures. However, no method can discover and quantify alternative structures with base pairs spanning arbitrarily long distances – an obstacle for studying viral, messenger, and long noncoding RNAs, which may form long-range base pairs. Here, we introduce the method of Structure Ensemble Ablation by Reverse Complement Hybridization with Mutational Profiling (SEARCH-MaP) and software for Structure Ensemble Inference by Sequencing, Mutation Identification, and Clustering of RNA (SEISMIC-RNA). We use SEARCH-MaP and SEISMIC-RNA to discover that the frameshift stimulating element of SARS coronavirus 2 base-pairs with another element 1 kilobase downstream in nearly half of RNA molecules, and that this structure competes with a pseudoknot that stimulates ribosomal frameshifting. Moreover, we identify long-range base pairs involving the frameshift stimulating element in other coronaviruses including SARS coronavirus 1 and transmissible gastroenteritis virus, and model the full genomic secondary structure of the latter. These findings suggest that long-range base pairs are common in coronaviruses and may regulate ribosomal frameshifting, which is essential for viral RNA synthesis. We anticipate that SEARCH-MaP will enable solving many RNA structure ensembles that have eluded characterization, thereby enhancing our general understanding of RNA structures and their functions. SEISMIC-RNA, software for analyzing mutational profiling data at any scale, could power future studies on RNA structure and is available on GitHub and the Python Package Index.

Understanding macromolecular structures of proteins and nucleic acids is critical for discerning their functions and biological roles. Advanced techniques—crystallography, nuclear magnetic resonance, and cryo–electron microscopy—have facilitated the determination of more than 180,000 protein structures, all cataloged in the Protein Data Bank. This comprehensive repository has been pivotal in developing deep learning algorithms for predicting protein structures directly from sequences. In contrast, RNA structure prediction has lagged and suffers from a scarcity of structural data. Here, we present the secondary structure models of 1098 primary microRNAs and 1456 human messenger RNA regions determined through chemical probing. We develop a deep learning architecture inspired from the Evoformer model of Alphafold and traditional architectures for secondary structure prediction. This model, eFold, was trained on our newly generated database and more than 300,000 secondary structures across multiple sources. We benchmark eFold on two challenging test sets of long and diverse RNA structures and show that our dataset and architecture contribute to increasing the prediction performance, compared to similar state-of-the-art methods. Together, our results reveal that merely expanding the database size is insufficient for generalization across families, whereas incorporating a greater diversity and complexity of RNA structures allows for enhanced model performance.