FoldamerDB: a database of peptidic foldamers

Interpretable molecular encodings and representations for machine learning tasks

Molecular encodings and their usage in machine learning models have demonstrated significant breakthroughs in biomedical applications, particularly in the classification of peptides and proteins. To this end, we propose a new encoding method: Interpretable Carbon-based Array of Neighborhoods (iCAN). Designed to address machine learning models' need for more structured and less flexible input, it captures the neighborhoods of carbon atoms in a counting array and improves the utility of the resulting encodings for machine learning models. The iCAN method provides interpretable molecular encodings and representations, enabling the comparison of molecular neighborhoods, identification of repeating patterns, and visualization of relevance heat maps for a given data set. When reproducing a large biomedical peptide classification study, it outperforms its predecessor encoding. When extended to proteins, it outperforms a lead structure-based encoding on 71% of the data sets. Our method offers interpretable encodings that can be applied to all organic molecules, including exotic amino acids, cyclic peptides, and larger proteins, making it highly versatile across various domains and data sets. This work establishes a promising new direction for machine learning in peptide and protein classification in biomedicine and healthcare, potentially accelerating advances in drug discovery and disease diagnosis.

Revealing the Effect of Stereocontrol on Intermolecular Interactions between Abiotic, Sequence-Defined Polyurethanes and a Ligand

The development of precision polymer synthesis has facilitated access to a diverse library of abiotic structures wherein chiral monomers are positioned at specific locations within macromolecular chains. These structures are anticipated to exhibit folding characteristics similar to those of biotic macromolecules and possess comparable functionalities. However, the extensive sequence space and numerous variables make selecting a sequence with the desired function challenging. Therefore, revealing sequence-function dependencies and developing practical tools are necessary to analyze their conformations and molecular interactions. In this study, we investigate the effect of stereochemistry, which dictates the spatial location of backbone and pendant groups, on the interaction between sequence-defined oligourethanes and bisphenol A ligands. Various methods are explored to analyze the receptor-like properties of model oligomers and the ligand. The accuracy of molecular dynamics simulations and experimental techniques is assessed to uncover the impact of discrete changes in stereochemical arrangements on the structures of the resulting complexes and their binding strengths. Detailed computational investigations providing atomistic details show that the formed complexes demonstrate significant structural diversity depending on the sequence of stereocenters, thus affecting the oligomer-ligand binding strength. Among the tested techniques, the fluorescence spectroscopy data, fitted to the Stern-Volmer equation, are consistently aligned with the calculations, thus validating the developed simulation methodology. The developed methodology opens a way to engineer the structure of sequence-defined oligomers with receptor-like functionality to explore their practical applications, e.g., as sensory materials.

Folding Coarse-Grained Oligomer Models with PyRosetta

Non-biological foldamers are a promising class of macromolecules that share similarities to classical biopolymers such as proteins and nucleic acids. Currently, designing novel foldamers is a non-trivial process, often involving many iterations of trial synthesis and characterization until folded structures are observed. In this work, we aim to tackle these foldamer design challenges using computational modeling techniques. We developed CG PyRosetta, an extension to the popular protein folding python package, PyRosetta, which introduces coarse-grained (CG) residues into PyRosetta, enabling the folding of toy CG foldamer models. Although these models are simplified, they can help explore overarching physical hypotheses about how oligomers can form. Through systematic variation of CG parameters in these models, we can investigate various folding hypotheses at the CG scale to inform the design process of new foldamer chemistries. In this study, we demonstrate CG PyRosetta's ability to identify minimum energy structures with a diverse structural search over a range of simple models, as well as two hypothesis-driven parameter scans investigating the effects of side-chain size and internal backbone angle on secondary structures. We are able to identify several types of secondary structures from single- and double-helices to sheet-like and knot-like structures. We show how side-chain size and backbone bond angle both play an important role in the structure and energetics of these toy models. Optimal side-chain sizes promote favorable packing of side chains, while specific backbone bond angles influence the specific helix type found in folded structures.

Helical Peptoid Ions in the Gas Phase: Thwarting the Charge Solvation Effect by H-Bond Compensation

Folding and unfolding processes are key aspects that should be mastered for the design of foldamer molecules for targeted applications. In contrast to the solution phase, in vacuo conditions represent a well-defined environment to analyze the intramolecular interactions that largely control the folding/unfolding dynamics. Ion mobility mass spectrometry coupled to theoretical modeling represents an efficient method to decipher the spatial structures of gaseous ions, including foldamers. However, charge solvation typically compacts the ion structure in the absence of strong stabilizing secondary interactions. This is the case in peptoids that are synthetic peptide regioisomers whose side chains are connected to the nitrogen atoms of the backbone instead of α-carbon as in peptides, thus implying the absence of H-bonds among the core units of the backbone. A recent work indeed reported that helical peptoids based on Nspe units formed in solution do not retain their secondary structure when transferred to the gas phase upon electrospray ionization (ESI). In this context, we demonstrate here that the helical structure of peptoids bearing (S)-N-(1-carboxy-2-phenylethyl) bulky side chains (Nscp) is largely preserved in the gas phase by the creation of a hydrogen bond network, induced by the presence of carboxylic moieties, that compensates for the charge solvation process.

Brilacidin Demonstrates Inhibition of SARS-CoV-2 in Cell Culture

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), the newly emergent causative agent of coronavirus disease-19 (COVID-19), has resulted in more than two million deaths worldwide since it was first detected in 2019. There is a critical global need for therapeutic intervention strategies that can be deployed to safely treat COVID-19 disease and reduce associated morbidity and mortality. Increasing evidence shows that both natural and synthetic antimicrobial peptides (AMPs), also referred to as Host Defense Proteins/Peptides (HDPs), can inhibit SARS-CoV-2, paving the way for the potential clinical use of these molecules as therapeutic options. In this manuscript, we describe the potent antiviral activity exerted by brilacidin-a de novo designed synthetic small molecule that captures the biological properties of HDPs-on SARS-CoV-2 in a human lung cell line (Calu-3) and a monkey cell line (Vero). These data suggest that SARS-CoV-2 inhibition in these cell culture models is likely to be a result of the impact of brilacidin on viral entry and its disruption of viral integrity. Brilacidin demonstrated synergistic antiviral activity when combined with remdesivir. Collectively, our data demonstrate that brilacidin exerts potent inhibition of SARS-CoV-2 against different strains of the virus in cell culture.

ConjuPepDB: a database of peptide-drug conjugates

Peptide–drug conjugates are organic molecules composed of (i) a small drug molecule, (ii) a peptide and (iii) a linker. The drug molecule is mandatory for the biological action, however, its efficacy can be enhanced by targeted delivery, which often also reduces unwanted side effects. For site-specificity the peptide part is mainly responsible. The linker attaches chemically the drug to the peptide, but it could also be biodegradable which ensures controlled liberation of the small drug. Despite the importance of the field, there is no public comprehensive database on these species. Herein we describe ConjuPepBD, a freely available, fully annotated and manually curated database of peptide drug conjugates. ConjuPepDB contains basic information about the entries, e.g. CAS number. Furthermore, it also implies their biomedical application and the type of chemical conjugation employed. It covers more than 1600 conjugates from ∼230 publications. The web-interface is user-friendly, intuitive, and useable on several devices, e.g. phones, tablets, PCs. The webpage allows the user to search for content using numerous criteria, chemical structure and a help page is also provided. Besides giving quick insight for newcomers, ConjuPepDB is hoped to be also helpful for researchers from various related fields. The database is accessible at: https://conjupepdb.ttk.hu/.

Membrane active Janus-oligomers of β3-peptides

Self-assembly of an acyclic β³-hexapeptide with alternating side chain chirality, into nanometer size oligomeric bundles showing membrane activity and hosting capacity for hydrophobic small molecules.

Self-assembling peptides offer a versatile set of tools for bottom-up construction of supramolecular biomaterials. Among these compounds, non-natural peptidic foldamers experience increased focus due to their structural variability and lower sensitivity to enzymatic degradation. However, very little is known about their membrane properties and complex oligomeric assemblies – key areas for biomedical and technological applications. Here we designed short, acyclic β³-peptide sequences with alternating amino acid stereoisomers to obtain non-helical molecules having hydrophilic charged residues on one side, and hydrophobic residues on the other side, with the N-terminus preventing formation of infinite fibrils. Our results indicate that these β-peptides form small oligomers both in water and in lipid bilayers and are stabilized by intermolecular hydrogen bonds. In the presence of model membranes, they either prefer the headgroup regions or they insert between the lipid chains. Molecular dynamics (MD) simulations suggest the formation of two-layered bundles with their side chains facing opposite directions when compared in water and in model membranes. Analysis of the MD calculations showed hydrogen bonds inside each layer, however, not between the layers, indicating a dynamic assembly. Moreover, the aqueous form of these oligomers can host fluorescent probes as well as a hydrophobic molecule similarly to e.g. lipid transfer proteins. For the tested, peptides the mixed chirality pattern resulted in similar assemblies despite sequential differences. Based on this, it is hoped that the presented molecular framework will inspire similar oligomers with diverse functionality.

Ultra-large chemical libraries for the discovery of high-affinity peptide binders

High-diversity genetically-encoded combinatorial libraries (108-1013 members) are a rich source of peptide-based binding molecules, identified by affinity selection. Synthetic libraries can access broader chemical space, but typically examine only ~ 106 compounds by screening. Here we show that in-solution affinity selection can be interfaced with nano-liquid chromatography-tandem mass spectrometry peptide sequencing to identify binders from fully randomized synthetic libraries of 108 members-a 100-fold gain in diversity over standard practice. To validate this approach, we show that binders to a monoclonal antibody are identified in proportion to library diversity, as diversity is increased from 106-108. These results are then applied to the discovery of p53-like binders to MDM2, and to a family of 3-19 nM-affinity, α/β-peptide-based binders to 14-3-3. An X-ray structure of one of these binders in complex with 14-3-3σ is determined, illustrating the role of β-amino acids in facilitating a key binding contact.

The 27th annual Nucleic Acids Research database issue and molecular biology database collection

The 2020 Nucleic Acids Research Database Issue contains 148 papers spanning molecular biology. They include 59 papers reporting on new databases and 79 covering recent changes to resources previously published in the issue. A further ten papers are updates on databases most recently published elsewhere. This issue contains three breakthrough articles: AntiBodies Chemically Defined (ABCD) curates antibody sequences and their cognate antigens; SCOP returns with a new schema and breaks away from a purely hierarchical structure; while the new Alliance of Genome Resources brings together a number of Model Organism databases to pool knowledge and tools. Major returning nucleic acid databases include miRDB and miRTarBase. Databases for protein sequence analysis include CDD, DisProt and ELM, alongside no fewer than four newcomers covering proteins involved in liquid-liquid phase separation. In metabolism and signaling, Pathway Commons, Reactome and Metabolights all contribute papers. PATRIC and MicroScope update in microbial genomes while human and model organism genomics resources include Ensembl, Ensembl genomes and UCSC Genome Browser. Immune-related proteins are covered by updates from IPD-IMGT/HLA and AFND, as well as newcomers VDJbase and OGRDB. Drug design is catered for by updates from the IUPHAR/BPS Guide to Pharmacology and the Therapeutic Target Database. The entire Database Issue is freely available online on the Nucleic Acids Research website (https://academic.oup.com/nar). The NAR online Molecular Biology Database Collection has been revised, updating 305 entries, adding 65 new resources and eliminating 125 discontinued URLs; so bringing the current total to 1637 databases. It is available at http://www.oxfordjournals.org/nar/database/c/.