VAE-Sim: A Novel Molecular Similarity Measure Based on a Variational Autoencoder

A Perspective on How Fibrinaloid Microclots and Platelet Pathology May be Applied in Clinical Investigations

Microscopy imaging has enabled us to establish the presence of fibrin(ogen) amyloid (fibrinaloid) microclots in a range of chronic, inflammatory diseases. Microclots may also be induced by a variety of purified substances, often at very low concentrations. These molecules include bacterial inflammagens, serum amyloid A, and the S1 spike protein of severe acute respiratory syndrome coronavirus 2. Here, we explore which of the properties of these microclots might be used to contribute to differential clinical diagnoses and prognoses of the various diseases with which they may be associated. Such properties include distributions in their size and number before and after the addition of exogenous thrombin, their spectral properties, the diameter of the fibers of which they are made, their resistance to proteolysis by various proteases, their cross-seeding ability, and the concentration dependence of their ability to bind small molecules including fluorogenic amyloid stains. Measuring these microclot parameters, together with microscopy imaging itself, along with methodologies like proteomics and imaging flow cytometry, as well as more conventional assays such as those for cytokines, might open up the possibility of a much finer use of these microclot properties in generative methods for a future where personalized medicine will be standard procedures in all clotting pathology disease diagnoses.

Designing molecules with autoencoder networks

Autoencoders are versatile tools in molecular informatics. These unsupervised neural networks serve diverse tasks such as data-driven molecular representation and constructive molecular design. This Review explores their algorithmic foundations and applications in drug discovery, highlighting the most active areas of development and the contributions autoencoder networks have made in advancing this field. We also explore the challenges and prospects concerning the utilization of autoencoders and the various adaptations of this neural network architecture in molecular design.

The use of deep learning technology in dance movement generation

The dance generated by the traditional music action matching and statistical mapping models is less consistent with the music itself. Moreover, new dance movements cannot be generated. A dance movement generation algorithm based on deep learning is designed to extract the mapping between sound and motion features to solve these problems. First, the sound and motion features are extracted from music and dance videos, and then, the model is built. In addition, a generator module, a discriminator module, and a self-encoder module are added to make the dance movement smoother and consistent with the music. The Pix2PixHD model is used to transform the dance pose sequence into a real version of the dance. Finally, the experiment takes the dance video on the network as the training data and trained 5,000 times. About 80% of the dance data are used as the training set and 20% as the test set. The experimental results show that Train, Valid, and Test values based on the Generator+Discriminator+Autoencoder model are 15.36, 17.19, and 19.12, respectively. The similarity between the generated dance sequence and the real dance sequence is 0.063, which shows that the proposed model can generate a dance more in line with the music. Moreover, the generated dance posture is closer to the real dance posture. The discussion has certain reference value for intelligent dance teaching, game field, cross-modal generation, and exploring the relationship between audio-visual information.

Modeling of the Crystallization Conditions for Organic Synthesis Product Purification Using Deep Learning

Crystallization is an important purification technique for solid products in a chemical laboratory. However, the correct selection of a solvent is important for the success of the procedure. In order to accelerate the solvent or solvent mixture search process, we offer an in silico alternative, i.e., a never previously demonstrated approach that can model the reaction mixture crystallization conditions which are invariant to the reaction type. The offered deep learning-based method is trained to directly predict the solvent labels used in the crystallization steps of the synthetic procedure. Our solvent label prediction task is a multi-label multi-class classification task during which the method must correctly choose one or several solvents from 13 possible examples. During the experimental investigation, we tested two multi-label classifiers (i.e., Feed-Forward and Long Short-Term Memory neural networks) applied on top of vectors. For the vectorization, we used two methods (i.e., extended-connectivity fingerprints and autoencoders) with various parameters. Our optimized technique was able to reach the accuracy of 0.870 ± 0.004 (which is 0.693 above the baseline) on the testing dataset. This allows us to assume that the proposed approach can help to accelerate manual R&D processes in chemical laboratories.

Feature Reduction for Molecular Similarity Searching Based on Autoencoder Deep Learning

The concept of molecular similarity has been commonly used in rational drug design, where structurally similar molecules are examined in molecular databases to retrieve functionally similar molecules. The most used conventional similarity methods used two-dimensional (2D) fingerprints to evaluate the similarity of molecules towards a target query. However, these descriptors include redundant and irrelevant features that might impact the performance of similarity searching methods. Thus, this study proposed a new approach for identifying the important features of molecules in chemical datasets based on the representation of the molecular features using Autoencoder (AE), with the aim of removing irrelevant and redundant features. The proposed approach experimented using the MDL Data Drug Report standard dataset (MDDR). Based on experimental findings, the proposed approach performed better than several existing benchmark similarity methods such as Tanimoto Similarity Method (TAN), Adapted Similarity Measure of Text Processing (ASMTP), and Quantum-Based Similarity Method (SQB). The results demonstrated that the performance achieved by the proposed approach has proven to be superior, particularly with the use of structurally heterogeneous datasets, where it yielded improved results compared to other previously used methods with the similar goal of improving molecular similarity searching.

Strategies for structure elucidation of small molecules based on LC–MS/MS data from complex biological samples

LC–MS/MS is a major analytical platform for metabolomics, which has become a recent hotspot in the research fields of life and environmental sciences. By contrast, structure elucidation of small molecules based on LC–MS/MS data remains a major challenge in the chemical and biological interpretation of untargeted metabolomics datasets. In recent years, several strategies for structure elucidation using LC–MS/MS data from complex biological samples have been proposed, these strategies can be simply categorized into two types, one based on structure annotation of mass spectra and for the other on retention time prediction. These strategies have helped many scientists conduct research in metabolite-related fields and are indispensable for the development of future tools. Here, we summarized the characteristics of the current tools and strategies for structure elucidation of small molecules based on LC–MS/MS data, and further discussed the directions and perspectives to improve the power of the tools or strategies for structure elucidation.

MassGenie: A Transformer-Based Deep Learning Method for Identifying Small Molecules from Their Mass Spectra

The 'inverse problem' of mass spectrometric molecular identification ('given a mass spectrum, calculate/predict the 2D structure of the molecule whence it came') is largely unsolved, and is especially acute in metabolomics where many small molecules remain unidentified. This is largely because the number of experimentally available electrospray mass spectra of small molecules is quite limited. However, the forward problem ('calculate a small molecule's likely fragmentation and hence at least some of its mass spectrum from its structure alone') is much more tractable, because the strengths of different chemical bonds are roughly known. This kind of molecular identification problem may be cast as a language translation problem in which the source language is a list of high-resolution mass spectral peaks and the 'translation' a representation (for instance in SMILES) of the molecule. It is thus suitable for attack using the deep neural networks known as transformers. We here present MassGenie, a method that uses a transformer-based deep neural network, trained on ~6 million chemical structures with augmented SMILES encoding and their paired molecular fragments as generated in silico, explicitly including the protonated molecular ion. This architecture (containing some 400 million elements) is used to predict the structure of a molecule from the various fragments that may be expected to be observed when some of its bonds are broken. Despite being given essentially no detailed nor explicit rules about molecular fragmentation methods, isotope patterns, rearrangements, neutral losses, and the like, MassGenie learns the effective properties of the mass spectral fragment and valency space, and can generate candidate molecular structures that are very close or identical to those of the 'true' molecules. We also use VAE-Sim, a previously published variational autoencoder, to generate candidate molecules that are 'similar' to the top hit. In addition to using the 'top hits' directly, we can produce a rank order of these by 'round-tripping' candidate molecules and comparing them with the true molecules, where known. As a proof of principle, we confine ourselves to positive electrospray mass spectra from molecules with a molecular mass of 500Da or lower, including those in the last CASMI challenge (for which the results are known), getting 49/93 (53%) precisely correct. The transformer method, applied here for the first time to mass spectral interpretation, works extremely effectively both for mass spectra generated in silico and on experimentally obtained mass spectra from pure compounds. It seems to act as a Las Vegas algorithm, in that it either gives the correct answer or simply states that it cannot find one. The ability to create and to 'learn' millions of fragmentation patterns in silico, and therefrom generate candidate structures (that do not have to be in existing libraries) directly, thus opens up entirely the field of de novo small molecule structure prediction from experimental mass spectra.

Intelligent host engineering for metabolic flux optimisation in biotechnology

Optimising the function of a protein of length N amino acids by directed evolution involves navigating a 'search space' of possible sequences of some 20N. Optimising the expression levels of P proteins that materially affect host performance, each of which might also take 20 (logarithmically spaced) values, implies a similar search space of 20P. In this combinatorial sense, then, the problems of directed protein evolution and of host engineering are broadly equivalent. In practice, however, they have different means for avoiding the inevitable difficulties of implementation. The spare capacity exhibited in metabolic networks implies that host engineering may admit substantial increases in flux to targets of interest. Thus, we rehearse the relevant issues for those wishing to understand and exploit those modern genome-wide host engineering tools and thinking that have been designed and developed to optimise fluxes towards desirable products in biotechnological processes, with a focus on microbial systems. The aim throughput is 'making such biology predictable'. Strategies have been aimed at both transcription and translation, especially for regulatory processes that can affect multiple targets. However, because there is a limit on how much protein a cell can produce, increasing kcat in selected targets may be a better strategy than increasing protein expression levels for optimal host engineering.

The Transporter-Mediated Cellular Uptake and Efflux of Pharmaceutical Drugs and Biotechnology Products: How and Why Phospholipid Bilayer Transport Is Negligible in Real Biomembranes

Over the years, my colleagues and I have come to realise that the likelihood of pharmaceutical drugs being able to diffuse through whatever unhindered phospholipid bilayer may exist in intact biological membranes in vivo is vanishingly low. This is because (i) most real biomembranes are mostly protein, not lipid, (ii) unlike purely lipid bilayers that can form transient aqueous channels, the high concentrations of proteins serve to stop such activity, (iii) natural evolution long ago selected against transport methods that just let any undesirable products enter a cell, (iv) transporters have now been identified for all kinds of molecules (even water) that were once thought not to require them, (v) many experiments show a massive variation in the uptake of drugs between different cells, tissues, and organisms, that cannot be explained if lipid bilayer transport is significant or if efflux were the only differentiator, and (vi) many experiments that manipulate the expression level of individual transporters as an independent variable demonstrate their role in drug and nutrient uptake (including in cytotoxicity or adverse drug reactions). This makes such transporters valuable both as a means of targeting drugs (not least anti-infectives) to selected cells or tissues and also as drug targets. The same considerations apply to the exploitation of substrate uptake and product efflux transporters in biotechnology. We are also beginning to recognise that transporters are more promiscuous, and antiporter activity is much more widespread, than had been realised, and that such processes are adaptive (i.e., were selected by natural evolution). The purpose of the present review is to summarise the above, and to rehearse and update readers on recent developments. These developments lead us to retain and indeed to strengthen our contention that for transmembrane pharmaceutical drug transport "phospholipid bilayer transport is negligible".

Prediction of Chromatography Conditions for Purification in Organic Synthesis Using Deep Learning

In this research, a process for developing normal-phase liquid chromatography solvent systems has been proposed. In contrast to the development of conditions via thin-layer chromatography (TLC), this process is based on the architecture of two hierarchically connected neural network-based components. Using a large database of reaction procedures allows those two components to perform an essential role in the machine-learning-based prediction of chromatographic purification conditions, i.e., solvents and the ratio between solvents. In our paper, we build two datasets and test various molecular vectorization approaches, such as extended-connectivity fingerprints, learned embedding, and auto-encoders along with different types of deep neural networks to demonstrate a novel method for modeling chromatographic solvent systems employing two neural networks in sequence. Afterward, we present our findings and provide insights on the most effective methods for solving prediction tasks. Our approach results in a system of two neural networks with long short-term memory (LSTM)-based auto-encoders, where the first predicts solvent labels (by reaching the classification accuracy of 0.950 ± 0.001) and in the case of two solvents, the second one predicts the ratio between two solvents (R2 metric equal to 0.982 ± 0.001). Our approach can be used as a guidance instrument in laboratories to accelerate scouting for suitable chromatography conditions.

FragNet, a Contrastive Learning-Based Transformer Model for Clustering, Interpreting, Visualizing, and Navigating Chemical Space

The question of molecular similarity is core in cheminformatics and is usually assessed via a pairwise comparison based on vectors of properties or molecular fingerprints. We recently exploited variational autoencoders to embed 6M molecules in a chemical space, such that their (Euclidean) distance within the latent space so formed could be assessed within the framework of the entire molecular set. However, the standard objective function used did not seek to manipulate the latent space so as to cluster the molecules based on any perceived similarity. Using a set of some 160,000 molecules of biological relevance, we here bring together three modern elements of deep learning to create a novel and disentangled latent space, viz transformers, contrastive learning, and an embedded autoencoder. The effective dimensionality of the latent space was varied such that clear separation of individual types of molecules could be observed within individual dimensions of the latent space. The capacity of the network was such that many dimensions were not populated at all. As before, we assessed the utility of the representation by comparing clozapine with its near neighbors, and we also did the same for various antibiotics related to flucloxacillin. Transformers, especially when as here coupled with contrastive learning, effectively provide one-shot learning and lead to a successful and disentangled representation of molecular latent spaces that at once uses the entire training set in their construction while allowing "similar" molecules to cluster together in an effective and interpretable way.

Semi-supervised Hierarchical Drug Embedding in Hyperbolic Space

Learning accurate drug representations is essential for tasks such as computational drug repositioning and prediction of drug side effects. A drug hierarchy is a valuable source that encodes knowledge of relations among drugs in a tree-like structure where drugs that act on the same organs, treat the same disease, or bind to the same biological target are grouped together. However, its utility in learning drug representations has not yet been explored, and currently described drug representations cannot place novel molecules in a drug hierarchy. Here, we develop a semi-supervised drug embedding that incorporates two sources of information: (1) underlying chemical grammar that is inferred from chemical structures of drugs and drug-like molecules (unsupervised) and (2) hierarchical relations that are encoded in an expert-crafted hierarchy of approved drugs (supervised). We use the Variational Auto-Encoder (VAE) framework to encode the chemical structures of molecules and use the drug-drug similarity information obtained from the hierarchy to induce the clustering of drugs in hyperbolic space. The hyperbolic space is amenable for encoding hierarchical relations. Both quantitative and qualitative results support that the learned drug embedding can accurately reproduce the chemical structure and recapitulate the hierarchical relations among drugs. Furthermore, our approach can infer the pharmacological properties of novel molecules by retrieving similar drugs from the embedding space. We demonstrate that our drug embedding can predict new uses and discover new side effects of existing drugs. We show that it significantly outperforms comparison methods in both tasks.

Deep learning and generative methods in cheminformatics and chemical biology: navigating small molecule space intelligently

The number of 'small' molecules that may be of interest to chemical biologists - chemical space - is enormous, but the fraction that have ever been made is tiny. Most strategies are discriminative, i.e. have involved 'forward' problems (have molecule, establish properties). However, we normally wish to solve the much harder generative or inverse problem (describe desired properties, find molecule). 'Deep' (machine) learning based on large-scale neural networks underpins technologies such as computer vision, natural language processing, driverless cars, and world-leading performance in games such as Go; it can also be applied to the solution of inverse problems in chemical biology. In particular, recent developments in deep learning admit the in silico generation of candidate molecular structures and the prediction of their properties, thereby allowing one to navigate (bio)chemical space intelligently. These methods are revolutionary but require an understanding of both (bio)chemistry and computer science to be exploited to best advantage. We give a high-level (non-mathematical) background to the deep learning revolution, and set out the crucial issue for chemical biology and informatics as a two-way mapping from the discrete nature of individual molecules to the continuous but high-dimensional latent representation that may best reflect chemical space. A variety of architectures can do this; we focus on a particular type known as variational autoencoders. We then provide some examples of recent successes of these kinds of approach, and a look towards the future.

Structural Similarities between Some Common Fluorophores Used in Biology, Marketed Drugs, Endogenous Metabolites, and Natural Products

It is known that at least some fluorophores can act as 'surrogate' substrates for solute carriers (SLCs) involved in pharmaceutical drug uptake, and this promiscuity is taken to reflect at least a certain structural similarity. As part of a comprehensive study seeking the 'natural' substrates of 'orphan' transporters that also serve to take up pharmaceutical drugs into cells, we have noted that many drugs bear structural similarities to natural products. A cursory inspection of common fluorophores indicates that they too are surprisingly 'drug-like', and they also enter at least some cells. Some are also known to be substrates of efflux transporters. Consequently, we sought to assess the structural similarity of common fluorophores to marketed drugs, endogenous mammalian metabolites, and natural products. We used a set of some 150 fluorophores along with standard fingerprinting methods and the Tanimoto similarity metric. Results: The great majority of fluorophores tested exhibited significant similarity (Tanimoto similarity > 0.75) to at least one drug, as judged via descriptor properties (especially their aromaticity, for identifiable reasons that we explain), by molecular fingerprints, by visual inspection, and via the "quantitative estimate of drug likeness" technique. It is concluded that this set of fluorophores does overlap with a significant part of both the drug space and natural products space. Consequently, fluorophores do indeed offer a much wider opportunity than had possibly been realised to be used as surrogate uptake molecules in the competitive or trans-stimulation assay of membrane transporter activities.

VAE-Sim: A Novel Molecular Similarity Measure Based on a Variational Autoencoder

Molecular similarity is an elusive but core "unsupervised" cheminformatics concept, yet different "fingerprint" encodings of molecular structures return very different similarity values, even when using the same similarity metric. Each encoding may be of value when applied to other problems with objective or target functions, implying that a priori none are "better" than the others, nor than encoding-free metrics such as maximum common substructure (MCSS). We here introduce a novel approach to molecular similarity, in the form of a variational autoencoder (VAE). This learns the joint distribution p(z|x) where z is a latent vector and x are the (same) input/output data. It takes the form of a "bowtie"-shaped artificial neural network. In the middle is a "bottleneck layer" or latent vector in which inputs are transformed into, and represented as, a vector of numbers (encoding), with a reverse process (decoding) seeking to return the SMILES string that was the input. We train a VAE on over six million druglike molecules and natural products (including over one million in the final holdout set). The VAE vector distances provide a rapid and novel metric for molecular similarity that is both easily and rapidly calculated. We describe the method and its application to a typical similarity problem in cheminformatics.