Analysis of the vibrational and sound spectrum of over 100,000 protein structures and application in sonification

Simultaneous Real and Momentum Space Electron Diffraction from a Fullerene Molecule

Plane-wave electrons undergo momentum transfer as they scatter off a target in overlapping spherical waves. The transferred momentum leads to target structural information to be encoded in angle and energy differential scattering. For symmetric, periodic, or structured targets this can engender diffraction in the electron intensity both in real and momentum space. With the example of elastic scattering from a C_{60} molecule we show these simultaneous diffraction signatures. Simulated angle-momentum diffractograms can be imaged in experiments with a two-dimensional detector and an energy-tunable electron gun. The result may inspire the invention of technology to extend scopes of electron diffraction studies, open a track of electron crystallography using the momentum-differential diffraction, and motivate research about controlling the time delay between the pump laser pulse and probe electron pulse by tuning the electron impact speed in ultrafast electron diffraction experiments.

Sound Matrix Shaping of Living Matter: From Macrosystems to Cell Microenvironment, Where Mitochondria Act as Energy Portals in Detecting and Processing Sound Vibrations

Vibration and sound are the shaping matrix of the entire universe. Everything in nature is shaped by energy vibrating and communicating through its own sound trail. Every cell within our body vibrates at defined frequencies, generating its peculiar "sound signature". Mitochondria are dynamic, energy-transforming, biosynthetic, and signaling organelles that actively transduce biological information. Novel research has shown that the mitochondrial function of mammalian cells can be modulated by various energetic stimuli, including sound vibrations. Regarding acoustic vibrations, definite types of music have been reported to produce beneficial impacts on human health. In very recent studies, the effects of different sound stimuli and musical styles on cellular function and mitochondrial activity were evaluated and compared in human cells cultured in vitro, investigating the underlying responsible molecular mechanisms. This narrative review will take a multilevel trip from macro to intracellular microenvironment, discussing the intimate vibrational sound activities shaping living matter, delving deeper into the molecular mechanisms underlying the sound modulation of biological systems, and mainly focusing our discussion on novel evidence showing the competence of mitochondria in acting as energy portals capable of sensing and transducing the subtle informational biofields of sound vibration.

Unsupervised cross-domain translation via deep learning and adversarial attention neural networks and application to music-inspired protein designs

Taking inspiration from nature about how to design materials has been a fruitful approach, used by humans for millennia. In this paper we report a method that allows us to discover how patterns in disparate domains can be reversibly related using a computationally rigorous approach, the AttentionCrossTranslation model. The algorithm discovers cycle- and self-consistent relationships and offers a bidirectional translation of information across disparate knowledge domains. The approach is validated with a set of known translation problems, and then used to discover a mapping between musical data-based on the corpus of note sequences in J.S. Bach's Goldberg Variations created in 1741-and protein sequence data-information sampled more recently. Using protein folding algorithms, 3D structures of the predicted protein sequences are generated, and their stability is validated using explicit solvent molecular dynamics. Musical scores generated from protein sequences are sonified and rendered into audible sound.

End-to-End Protein Normal Mode Frequency Predictions Using Language and Graph Models and Application to Sonification

The prediction of mechanical and dynamical properties of proteins is an important frontier, especially given the greater availability of proteins structures. Here we report a series of models that provide end-to-end predictions of nanodynamical properties of proteins, focused on high-throughput normal mode predictions directly from the amino acid sequence. Using neural network models within the family of Natural Language Processing and graph-based methods, we offer atomistically based mechanistic predictions of key protein mechanical features. The models include an end-to-end long short-term memory (LSTM) model, an end-to-end transformer model, a graph-based transformer model, and an equivariant graph neural network. All four models show exceptional performance, with the graph-based transformer architecture offering the best results but at the cost of requiring a graph structure as input. Conversely, the LSTM and transformer models offer end-to-end sequence-to-property prediction capabilities, providing efficient avenues for protein engineering, analysis, and design. We compare our results against published data based on a Principal Neighborhood Aggregation graph neural network, revealing that the transformer model offers better performance while also being able to predict a large set of the first 64 normal mode frequencies, simultaneously. The use of the end-to-end transformer model may facilitate other downstream applications through the use of transfer learning, and it offers a comprehensive prediction of dynamical properties without any structural knowledge, directly from the amino acid sequence. We demonstrate a potential application in scientific sonification, where the normal mode frequencies are transposed to generate audible signals for a detailed analysis of subtle changes of protein sequences.

Multiscale Modeling at the Interface of Molecular Mechanics and Natural Language through Attention Neural Networks

Humans are continually bombarded with massive amounts of data. To deal with this influx of information, we use the concept of attention in order to perceive the most relevant input from vision, hearing, touch, and others. Thereby, the complex ensemble of signals is used to generate output by querying the processed data in appropriate ways. Attention is also the hallmark of the development of scientific theories, where we elucidate which parts of a problem are critical, often expressed through differential equations. In this Account we review the emergence of attention-based neural networks as a class of approaches that offer many opportunities to describe materials across scales and modalities, including how universal building blocks interact to yield a set of material properties. In fact, the self-assembly of hierarchical, structurally complex, and multifunctional biomaterials remains a grand challenge in modeling, theory, and experiment. Expanding from the process by which material building blocks physically interact to form a type of material, in this Account we view self-assembly as both the functional emergence of properties from interacting building blocks as well as the physical process by which elementary building blocks interact and yield structure and, thereby, functions. This perspective, integrated through the theory of materiomics, allows us to solve multiscale problems with a first-principles-based computational approach based on attention-based neural networks that transform information to feature to property while providing a flexible modeling approach that can integrate theory, simulation, and experiment. Since these models are based on a natural language framework, they offer various benefits including incorporation of general domain knowledge via general-purpose pretraining, which can be accomplished without labeled data or large amounts of lower-quality data. Pretrained models then offer a general-purpose platform that can be fine-tuned to adapt these models to make specific predictions, often with relatively little labeled data. The transferrable power of the language-based modeling approach realizes a neural olog description, where mathematical categorization is learned by multiheaded attention, without domain knowledge in its formulation. It can hence be applied to a range of complex modeling tasks─such as physical field predictions, molecular properties, or structure predictions, all using an identical formulation. This offers a complementary modeling approach that is already finding numerous applications, with great potential to solve complex assembly problems, enabling us to learn, build, and utilize functional categorization of how building blocks yield a range of material functions. In this Account, we demonstrate the approach in various application areas, including protein secondary structure prediction and prediction of normal-mode frequencies as well as predicting mechanical fields near cracks. Unifying these diverse problem areas is the building block approach, where the models are based on a universally applicable platform that offers benefits ranging from transferability, interpretability, and cross-domain pollination of knowledge as exemplified through a transformer model applied to predict how musical compositions infer de novo protein structures. We discuss future potentialities of this approach for a variety of material phenomena across scales, including the use in multiparadigm modeling schemes.

Rapid prediction of protein natural frequencies using graph neural networks

Natural vibrational frequencies of proteins help to correlate functional shifts with sequence or geometric variations that lead to negligible changes in protein structures, such as point mutations related to disease lethality or medication effectiveness. Normal mode analysis is a well-known approach to accurately obtain protein natural frequencies. However, it is not feasible when high-resolution protein structures are not available or time consuming to obtain. Here we provide a machine learning model to directly predict protein frequencies from primary amino acid sequences and low-resolution structural features such as contact or distance maps. We utilize a graph neural network called principal neighborhood aggregation, trained with the structural graphs and normal mode frequencies of more than 34 000 proteins from the protein data bank. combining with existing contact/distance map prediction tools, this approach enables an end-to-end prediction of the frequency spectrum of a protein given its primary sequence.

We present a computational framework based on graph neural networks (GNNs) to predict the natural frequencies of proteins from primary amino acid sequences and contact/distance maps.

Bioinspired translation of classical music intode novoprotein structures using deep learning and molecular modeling

Architected biomaterials, as well as sound and music, are constructed from small building blocks that are assembled across time- and length-scales. Here we present a novel deep learning-enabled integrated algorithmic workflow to merge the two concepts for radical discovery ofde novoprotein materials, exploiting musical creativity as the foundation, and extrapolating through a recursive method to increase protein complexity by successively injecting protein chemistry into the process. Indeed, music is one of the few universal expressions that can create bridges between cultures, find associations between seemingly unrelated concepts, and can be used as a novel way to generate bio-inspired designs that derive functions from the imaginations of the creative mind. Earlier work has offered a pathway to convert proteins into sound, and sound into proteins. Here we build on this paradigm and translate a piece of classical music into matter. Based on Bach's Goldberg variations, we offer a series of case studies to convert the musical data imagined by the composer into protein design, and folded into a 3D structure using deep learning. The quest we seek to address is to identify semblances, or memories, or information content in such musical creation, that offers new insights into pattern relationships between distinct manifestations of information. Using basic local alignment search tool analysis, we find that several fragments of the new proteins display similarities to existing protein sequences found in proteobacteria among other organisms, especially in regions of low complexity and repetitive motifs. The resulting protein forms the basis for iterative musical composition, and an evolutionary paradigm that defines a variational pathway for melodic development, complementing conventional creative or mathematical methods. This paper broadens the concept of what is understood as bio-inspiration to include a broad array of systems created by humans, animals, or other natural mechanisms.

Artificial intelligence-driven drug repurposing and structural biology for SARS-CoV-2

It has been said that COVID-19 is a generational challenge in many ways. But, at the same time, it becomes a catalyst for collective action, innovation, and discovery. Realizing the full potential of artificial intelligence (AI) for structure determination of unknown proteins and drug discovery are some of these innovations. Potential applications of AI include predicting the structure of the infectious proteins, identifying drugs that may be effective in targeting these proteins, and proposing new chemical compounds for further testing as potential drugs. AI and machine learning (ML) allow for rapid drug development including repurposing existing drugs. Algorithms were used to search for novel or approved antiviral drugs capable of inhibiting SARS-CoV-2. This paper presents a survey of AI and ML methods being used in various biochemistry of SARS-CoV-2, from structure to drug development, in the fight against the deadly COVID-19 pandemic. It is envisioned that this study will provide AI/ML researchers and the wider community an overview of the current status of AI applications particularly in structural biology, drug repurposing, and development, and motivate researchers in harnessing AI potentials in the fight against COVID-19.

Comparative Analysis of Nanomechanical Features of Coronavirus Spike Proteins and Correlation with Lethality and Infection Rate

The novel coronavirus disease 2019 (COVID-19) has spread rapidly around the world. Its causative virus, severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), enters human cells through the physical interaction between the receptor-binding domain (RBD) of its spike protein and the human cell receptor ACE2. Here, we provide a novel way of understanding coronavirus spike proteins, connecting their nanomechanical features, specifically their vibrational spectrum and quantitative measures of mobility, with virus lethality and infection rate. The key result of our work is that both the overall flexibility of upward RBD and the mobility ratio of RBDs in different conformations represent two significant factors that show a positive scaling with virus lethality and an inverse correlation with the infection rate. Our analysis shows that epidemiological virus properties can be linked directly to pure nanomechanical, vibrational aspects, offering an alternative way of screening new viruses and mutations, and potentially exploring novel ways to prevent infections from occurring.

Sonification based de novo protein design using artificial intelligence, structure prediction, and analysis using molecular modeling

We report the use of a deep learning model to design de novo proteins, based on the interplay of elementary building blocks via hierarchical patterns. The deep neural network model is based on translating protein sequences and structural information into a musical score that features different pitches for each of the amino acids, and variations in note length and note volume reflecting secondary structure information and information about the chain length and distinct protein molecules. We train a deep learning model whose architecture is composed of several long short-term memory units from data consisting of musical representations of proteins classified by certain features, focused here on alpha-helix rich proteins. Using the deep learning model, we then generate de novo musical scores and translate the pitch information and chain lengths into sequences of amino acids. We use a Basic Local Alignment Search Tool to compare the predicted amino acid sequences against known proteins, and estimate folded protein structures using the Optimized protein fold RecognitION method (ORION) and MODELLER. We find that the method proposed here can be used to design de novo proteins that do not exist yet, and that the designed proteins fold into specified secondary structures. We validate the newly predicted protein by molecular dynamics equilibration in explicit water and subsequent characterization using a normal mode analysis. The method provides a tool to design novel protein materials that could find useful applications as materials in biology, medicine, and engineering.

Wave Propagation and Energy Dissipation in Collagen Molecules

Collagen is the key protein of connective tissue (i.e., skin, tendons and ligaments, and cartilage, among others), accounting for 25-35% of the whole-body protein content and conferring mechanical stability. This protein is also a fundamental building block of bone because of its excellent mechanical properties together with carbonated hydroxyapatite minerals. Although the mechanical resilience and viscoelasticity have been studied both in vitro and in vivo from the molecular to tissue level, wave propagation properties and energy dissipation have not yet been deeply explored, in spite of being crucial to understanding the vibration dynamics of collagenous structures (e.g., eardrum, cochlear membranes) upon impulsive loads. By using a bottom-up atomistic modeling approach, here we study a collagen peptide under two distinct impulsive displacement loads, including longitudinal and transversal inputs. Using a one-dimensional string model as a model system, we investigate the roles of hydration and load direction on wave propagation along the collagen peptide and the related energy dissipation. We find that wave transmission and energy-dissipation strongly depend on the loading direction. Also, the hydrated collagen peptide can dissipate five times more energy than dehydrated one. Our work suggests a distinct role of collagen in term of wave transmission of different tissues such as tendon and eardrum. This study can step toward understanding the mechanical behavior of collagen upon transient loads, impact loading and fatigue, and designing biomimetic and bioinspired materials to replace specific native tissues such as the tympanic membrane.

Machine learning model for fast prediction of the natural frequencies of protein molecules

Natural vibrations and resonances are intrinsic features of protein structures and enable differentiation of one structure from another. These nanoscale features are important to help to understand the dynamics of a protein molecule and identify the effects of small sequence or other geometric alterations that may not cause significant visible structural changes, such as point mutations associated with disease or drug design. Although normal mode analysis provides a powerful way to accurately extract the natural frequencies of a protein, it must meet several critical conditions, including availability of high-resolution structures, availability of good chemical force fields and memory-intensive large-scale computing resources. Here, we study the natural frequency of over 100 000 known protein molecular structures from the Protein Data Bank and use this dataset to carefully investigate the correlation between their structural features and these natural frequencies by using a machine learning model composed of a Feedforward Neural Network made of four hidden layers that predicts the natural frequencies in excellent agreement with full-atomistic normal mode calculations, but is significantly more computationally efficient. In addition to the computational advance, we demonstrate that this model can be used to directly obtain the natural frequencies by merely using five structural features of protein molecules as predictor variables, including the largest and smallest diameter, and the ratio of amino acid residues with alpha-helix, beta strand and 3–10 helix domains. These structural features can be either experimentally or computationally obtained, and do not require a full-atomistic model of a protein of interest. This method is helpful in predicting the absorption and resonance functions of an unknown protein molecule without solving its full atomic structure.

Natural vibrations and resonances are intrinsic features of protein structures and can be learnt from existing structures.

Sounds interesting: can sonification help us design new proteins?

Introduction: The practice of turning scientific data into music, a practice known as sonification, is a growing field. Driven by analogies between the hierarchical structures of proteins and many forms of music, multiple attempts of mapping proteins to music have been made. Previous works have either worked at a low level, mapping amino acid to notes, or at a higher level, using the overall structure as a basis for composition.Areas covered: We report a comprehensive mapping strategy that encompasses the encoding of the geometry of proteins, in addition to the amino acid sequence and secondary structure information. This leads to a piece of music that is both more complete and closely linked to the original protein. By using this mapping, we can invert the process and map music to proteins, retrieving not only the amino acid sequence but also the secondary structure and folding from musical data.Expert opinion: We can train a machine learning model on 'protein music' to generate new music that can be translated to new proteins. By selecting proper datasets and conditioning parameters on the generative model, we could tune de novo proteins with high level parameters to achieve certain protein design features.

A Self-Consistent Sonification Method to Translate Amino Acid Sequences into Musical Compositions and Application in Protein Design Using Artificial Intelligence

We report a self-consistent method to translate amino acid sequences into audible sound, use the representation in the musical space to train a neural network, and then apply it to generate protein designs using artificial intelligence (AI). The sonification method proposed here uses the normal mode vibrations of the amino acid building blocks of proteins to compute an audible representation of each of the 20 natural amino acids, which is fully defined by the overlay of its respective natural vibrations. The vibrational frequencies are transposed to the audible spectrum following the musical concept of transpositional equivalence, playing or writing music in a way that makes it sound higher or lower in pitch while retaining the relationships between tones or chords played. This transposition method ensures that the relative values of the vibrational frequencies within each amino acid and among different amino acids are retained. The characteristic frequency spectrum and sound associated with each of the amino acids represents a type of musical scale that consists of 20 tones, the "amino acid scale". To create a playable instrument, each tone associated with the amino acids is assigned to a specific key on a piano roll, which allows us to map the sequence of amino acids in proteins into a musical score. To reflect higher-order structural details of proteins, the volume and duration of the notes associated with each amino acid are defined by the secondary structure of proteins, computed using DSSP and thereby introducing musical rhythm. We then train a recurrent neural network based on a large set of musical scores generated by this sonification method and use AI to generate musical compositions, capturing the innate relationships between amino acid sequence and protein structure. We then translate the de novo musical data generated by AI into protein sequences, thereby obtaining de novo protein designs that feature specific design characteristics. We illustrate the approach in several examples that reflect the sonification of protein sequences, including multihour audible representations of natural proteins and protein-based musical compositions solely generated by AI. The approach proposed here may provide an avenue for understanding sequence patterns, variations, and mutations and offers an outreach mechanism to explain the significance of protein sequences. The method may also offer insight into protein folding and understanding the context of the amino acid sequence in defining the secondary and higher-order folded structure of proteins and could hence be used to detect the effects of mutations through sound.