Sequence-based drug design as a concept in computational drug design.

Drug development based on target proteins has been a successful approach in recent decades. However, the conventional structure-based drug design (SBDD) pipeline is a complex, human-engineered process with multiple independently optimized steps. Here, we propose a sequence-to-drug concept for computational drug design based on protein sequence information by end-to-end differentiable learning. We validate this concept in three stages. First, we design TransformerCPI2.0 as a core tool for the concept, which demonstrates generalization ability across proteins and compounds. Second, we interpret the binding knowledge that TransformerCPI2.0 learned. Finally, we use TransformerCPI2.0 to discover new hits for challenging drug targets, and identify new target for an existing drug based on an inverse application of the concept. Overall, this proof-of-concept study shows that the sequence-to-drug concept adds a perspective on drug design. It can serve as an alternative method to SBDD, particularly for proteins that do not yet have high-quality 3D structures available.

Targeting UHRF1-SAP30-MXD4 axis for leukemia initiating cell eradication in myeloid leukemia.

Aberrant self-renewal of leukemia initiation cells (LICs) drives aggressive acute myeloid leukemia (AML). Here, we report that UHRF1, an epigenetic regulator that recruits DNMT1 to methylate DNA, is highly expressed in AML and predicts poor prognosis. UHRF1 is required for myeloid leukemogenesis by maintaining self-renewal of LICs. Mechanistically, UHRF1 directly interacts with Sin3A-associated protein 30 (SAP30) through two critical amino acids, G572 and F573 in its SRA domain, to repress gene expression. Depletion of UHRF1 or SAP30 derepresses an important target gene, MXD4, which encodes a MYC antagonist, and leads to suppression of leukemogenesis. Further knockdown of MXD4 can rescue the leukemogenesis by activating the MYC pathway. Lastly, we identified a UHRF1 inhibitor, UF146, and demonstrated its significant therapeutic efficacy in the myeloid leukemia PDX model. Taken together, our study reveals the mechanisms for altered epigenetic programs in AML and provides a promising targeted therapeutic strategy against AML.

Prediction of serious eye damage or eye irritation potential of compounds via consensus labelling models and active learning models based on uncertainty strategies.

Serious eye damage and eye irritation have been authenticated to be significant human health issues in various fields such as ophthalmic pharmaceuticals. Due to the shortcomings of traditional animal testing methods, in silico methods have advanced to study eye toxicity. The models for predicting serious eye damage and eye irritation potential of compounds were developed using 2299 and 5214 compounds, respectively. The 40 global single models and 40 local models were developed by combining 5 molecular description methods and 4 machine learning methods. The 40 active learning models were developed by adopting uncertainty-based active learning strategies and taking local models as initial models. The 110 global consensus models based on 40 global single models were developed using a consensus strategy. Active learning models and global consensus models performed high prediction accuracy. The test accuracy of the best serious eye damage model and eye irritation model reached 0.972 and 0.959, respectively. The applicability domains for all models were calculated to verify the rationality of prediction effect. In addition, 8 structural alerts probably causing serious eye damage or eye irritation were sought out. The prediction models and structural alerts contributed to providing hazard identification and assessing chemical safety.

Discovery of Pyrazolo[3,4-d]pyridazinone Derivatives as Selective DDR1 Inhibitors via Deep Learning Based Design, Synthesis, and Biological Evaluation.

Alterations of discoidin domain receptor1 (DDR1) may lead to increased production of inflammatory cytokines, making DDR1 an attractive target for inflammatory bowel disease (IBD) therapy. A scaffold-based molecular design workflow was established and performed by integrating a deep generative model, kinase selectivity screening and molecular docking, leading to a novel DDR1 inhibitor compound 2, which showed potent DDR1 inhibition profile (IC50 = 10.6 ± 1.9 nM) and excellent selectivity against a panel of 430 kinases (S (10) = 0.002 at 0.1 µM). Compound 2 potently inhibited the expression of pro-inflammatory cytokines and DDR1 autophosphorylation in cells, and it also demonstrated promising oral therapeutic effect in a dextran sulfate sodium (DSS)-induced mouse colitis model.

Drug target inference by mining transcriptional data using a novel graph convolutional network framework.

A fundamental challenge that arises in biomedicine is the need to characterize compounds in a relevant cellular context in order to reveal potential on-target or off-target effects. Recently, the fast accumulation of gene transcriptional profiling data provides us an unprecedented opportunity to explore the protein targets of chemical compounds from the perspective of cell transcriptomics and RNA biology. Here, we propose a novel Siamese spectral-based graph convolutional network (SSGCN) model for inferring the protein targets of chemical compounds from gene transcriptional profiles. Although the gene signature of a compound perturbation only provides indirect clues of the interacting targets, and the biological networks under different experiment conditions further complicate the situation, the SSGCN model was successfully trained to learn from known compound-target pairs by uncovering the hidden correlations between compound perturbation profiles and gene knockdown profiles. On a benchmark set and a large time-split validation dataset, the model achieved higher target inference accuracy as compared to previous methods such as Connectivity Map. Further experimental validations of prediction results highlight the practical usefulness of SSGCN in either inferring the interacting targets of compound, or reversely, in finding novel inhibitors of a given target of interest.

Design, synthesis and biological evaluation of pyrazolo[3,4-d]pyridazinone derivatives as covalent FGFR inhibitors.

Fibroblast growth factor receptors (FGFRs) have emerged as promising targets for anticancer therapy. In this study, we synthesized and evaluated the biological activity of 66 pyrazolo[3,4-d]pyridazinone derivatives. Kinase inhibition, cell proliferation, and whole blood stability assays were used to evaluate their activity on FGFR, allowing us to explore structure-activity relationships and thus to gain understanding of the structural requirements to modulate covalent inhibitors' selectivity and reactivity. Among them, compound 10h exhibited potent enzymatic activity against FGFR and remarkably inhibited proliferation of various cancer cells associated with FGFR dysregulation, and suppressed FGFR signaling pathway in cancer cells by the immunoblot analysis. Moreover, 10h displayed highly potent antitumor efficacy (TGI = 91.6%, at a dose of 50 mg/kg) in the FGFR1-amplified NCI-H1581 xenograft model.

DrugSpaceX: a large screenable and synthetically tractable database extending drug space.

One of the most prominent topics in drug discovery is efficient exploration of the vast drug-like chemical space to find synthesizable and novel chemical structures with desired biological properties. To address this challenge, we created the DrugSpaceX (https://drugspacex.simm.ac.cn/) database based on expert-defined transformations of approved drug molecules. The current version of DrugSpaceX contains >100 million transformed chemical products for virtual screening, with outstanding characteristics in terms of structural novelty, diversity and large three-dimensional chemical space coverage. To illustrate its practical application in drug discovery, we used a case study of discoidin domain receptor 1 (DDR1), a kinase target implicated in fibrosis and other diseases, to show DrugSpaceX performing a quick search of initial hit compounds. Additionally, for ligand identification and optimization purposes, DrugSpaceX also provides several subsets for download, including a 10% diversity subset, an extended drug-like subset, a drug-like subset, a lead-like subset, and a fragment-like subset. In addition to chemical properties and transformation instructions, DrugSpaceX can locate the position of transformation, which will enable medicinal chemists to easily integrate strategy planning and protection design.

TransformerCPI: improving compound-protein interaction prediction by sequence-based deep learning with self-attention mechanism and label reversal experiments.

MOTIVATION: Identifying compound-protein interaction (CPI) is a crucial task in drug discovery and chemogenomics studies, and proteins without three-dimensional structure account for a large part of potential biological targets, which requires developing methods using only protein sequence information to predict CPI. However, sequence-based CPI models may face some specific pitfalls, including using inappropriate datasets, hidden ligand bias and splitting datasets inappropriately, resulting in overestimation of their prediction performance. RESULTS: To address these issues, we here constructed new datasets specific for CPI prediction, proposed a novel transformer neural network named TransformerCPI, and introduced a more rigorous label reversal experiment to test whether a model learns true interaction features. TransformerCPI achieved much improved performance on the new experiments, and it can be deconvolved to highlight important interacting regions of protein sequences and compound atoms, which may contribute chemical biology studies with useful guidance for further ligand structural optimization. AVAILABILITY AND IMPLEMENTATION: https://github.com/lifanchen-simm/transformerCPI.

Revisiting Aldehyde Oxidase Mediated Metabolism in Drug-like Molecules: An Improved Computational Model.

Aldehyde oxidase (AOX) is a drug metabolizing molybdo-flavoenzyme that has gained increasing attention because of contribution to the biotransformation in phase I metabolism of xenobiotics. Unfortunately, the intra- and interspecies variations in AOX activity and lack of reliable and predictive animal models make evaluation of AOX-catalyzed metabolism prone to be misleading. In this study, we developed an improved computational model integrating both atom-level and molecule-level features to predict whether a drug-like molecule is a potential human AOX (hAOX) substrate and to identify the corresponding sites of metabolism. Additionally, we combined the proposed computational strategy and in vitro experiments for evaluating the metabolic property of a series of epigenetic-related drug candidates still in the early stage of development. In summary, this study provides an improved strategy to evaluate the liability of molecules toward hAOX and offers useful information for accelerating the drug design and optimization stage.

Deep Learning Enhancing Kinome-Wide Polypharmacology Profiling: Model Construction and Experiment Validation.

The kinome-wide virtual profiling of small molecules with high-dimensional structure-activity data is a challenging task in drug discovery. Here, we present a virtual profiling model against a panel of 391 kinases based on large-scale bioactivity data and the multitask deep neural network algorithm. The obtained model yields excellent internal prediction capability with an auROC of 0.90 and consistently outperforms conventional single-task models on external tests, especially for kinases with insufficient activity data. Moreover, more rigorous experimental validations including 1410 kinase-compound pairs showed a high-quality average auROC of 0.75 and confirmed many novel predicted "off-target" activities. Given the verified generalizability, the model was further applied to various scenarios for depicting the kinome-wide selectivity and the association with certain diseases. Overall, the computational model enables us to create a comprehensive kinome interaction network for designing novel chemical modulators or drug repositioning and is of practical value for exploring previously less studied kinases.

Discovery and characterization of natural products as novel indoleamine 2,3-dioxygenase 1 inhibitors through high-throughput screening.

Indoleamine 2,3-dioxygenase 1 (IDO1) is emerging as a promising therapeutic target for the treatment of malignant tumors characterized by dysregulated tryptophan metabolism. However, the antitumor efficacy of existing small-molecule IDO1 inhibitors is still unsatisfactory, and the underlying mechanism remains largely undefined. To identify novel IDO1 inhibitors, an in-house natural product library of 2000 natural products was screened for inhibitory activity against recombinant human IDO1. High-throughput fluorescence-based screening identified 79 compounds with inhibitory activity > 30% at 20 µM. Nine natural products were further confirmed to inhibit IDO1 activity by > 30% using Ehrlich's reagent reaction. Compounds 2, 7, and 8 were demonstrated to inhibit IDO1 activity in a cellular context. Compounds 2 and 7 were more potent against IDO1 than TDO2 in the enzymatic assay. The kinetic studies showed that compound 2 exhibited noncompetitive inhibition, whereas compounds 7 and 8 were graphically well matched with uncompetitive inhibition. Compounds 7 and 8 were found to bind to the ferric-IDO1 enzyme. Docking stimulations showed that the naphthalene ring of compound 8 formed "T-shaped" π-π interactions with Phe-163 and that the 6-methyl-naphthalene group formed additional hydrophobic interactions with IDO1. Compound 8 was identified as a derivative of tanshinone, and preliminary SAR analysis indicated that tanshinone derivatives may be promising hits for the development of IDO1 inhibitors. This study provides new clues for the discovery of IDO1/TDO2 inhibitors with novel scaffolds.

Application of Subwindow Factor Analysis and Mass Spectral information for accurate alignment of non-targeted metabolic profiling.

The peak shifts may lead to an incorrect statistical result for nontargeted metabolomics profiling, such as classification and discrimination in pattern recognition. In the paper, a more accurate alignment algorithm is developed based on Subwindow Factor Analysis and Mass Spectral information (SFA-MS). Compared with other methods, this new algorithm aligns the peaks more accurately without changing their shapes, especially for the overlapping peak clusters. To begin, the Continuous Wavelet Transform with Haar wavelet as the mother wavelet (Haar CWT) is used to determine the position and width of peaks. On this basis, the candidate drift points are confirmed by Fast Fourier Transform (FFT) cross correlation. Furthermore, the MS fitting degree of the common components between the reference chromatogram and the raw chromatogram is determined by the Subwindow Factor Analysis (SFA). When the MS information between reference and raw peaks is identical, the corresponding moving points are the optimum shifts. It is remarkable that all the peaks are moved through linear interpolation in the non-peak parts, so that the aligned chromatograms remain unchanged. The SFA-MS algorithm was implemented in the Matlab language and is available as an open source package.

Multi-analytical strategy for unassigned peaks using physical/mathematical separation, fragmental rules and retention index prediction: An example of sesquiterpene metabolites characterization in Cyperus rotundus.

Comprehensive two-dimensional gas chromatography- mass spectrometry (GC × GC-qMS) can provide powerful physical separation, signal enhancement, and spectral identification for analytes in complex samples. Unassigned peaks are commonly presented in the untargeted profile after a single run with EI-MS spectral matching and retention index (RI) confirmation. The procedure proposed in this work can be applied as a general method for suggesting or narrowing down the candidates of unassigned GC × GC-qMS peaks. To begin, peak purity detection and chemometric resolution are employed to acquire pure mass spectra. In addition, the fragmental rules and in-silico spectra from structures are available for annotating certain unassigned peaks with reference spectra that are not observed in commercial databases. Furthermore, the procedure proposed in this work allows for in silico RI calculation by means of random forest (RF) analysis based on the retention data under the same chromatographic conditions. The calculated RIs can aid in analysis when the RI information of peaks of interest is not available in retention data libraries. Using the proposed strategy, certain unassigned peaks can be attributed to sesquiterpene metabolites in an in-house database for Cyperus rotundus.

A modified multiscale peak alignment method combined with trilinear decomposition to study the volatile/heat-labile components in Ligusticum chuanxiong Hort - Cyperus rotundus rhizomes by HS-SPME-GC/MS.

Head Space/Solid Phase Micro-Extraction (HS-SPME) coupled with Gas Chromatography/Mass Spectrometer (GC/MS) was used to determine the volatile/heat-labile components in Ligusticum chuanxiong Hort - Cyperus rotundus rhizomes. Facing co-eluting peaks in k samples, a trilinear structure was reconstructed to obtain the second-order advantage. The retention time (RT) shift with multi-channel detection signals for different samples has been vital in maintaining the trilinear structure, thus a modified multiscale peak alignment (mMSPA) method was proposed in this paper. The peak position and peak width of representative ion profile were firstly detected by mMSPA using Continuous Wavelet Transform with Haar wavelet as the mother wavelet (Haar CWT). Then, the raw shift was confirmed by Fast Fourier Transform (FFT) cross correlation calculation. To obtain the optimal shift, Haar CWT was again used to detect the subtle deviations and be amalgamated in calculation. Here, to ensure there is no peaks shape alternation, the alignment was performed in local domains of data matrices, and all data points in the peak zone were moved via linear interpolation in non-peak parts. Finally, chemical components of interest in Ligusticum chuanxiong Hort - Cyperus rotundus rhizomes were analyzed by HS-SPME-GCMS and mMSPA-alternating trilinear decomposition (ATLD) resolution. As a result, the concentration variation between herbs and their pharmaceutical products can provide a scientific basic for the quality standard establishment of traditional Chinese medicines.

Accurate recognition and feature qualify for flavonoid extracts from Liang-wai Gan Cao by liquid chromatography-high resolution-mass spectrometry and computational MS/MS fragmentation.

In this study, Liquid Chromatography (LC) separation combined with quadrupole-Time-Of-Flight Mass Spectrometry (qTOF-MS) detection was used to analyze the characteristic ions of the flavonoids from Liang-wai Gan Cao (Radix Glycyrrhizae uralensis). First, accurate mass measurement and isotope curve optimization could provide reliable molecular prediction after noise deduction, baseline calibration and "ghost peak recognition". Thus, some spectral features in the LC-MS data could be clearly explained. Secondly, the chemical structure of flavonoids was deduced by MS/MS fragment ions, and the in-silico spectra by MS-FINDER program provided strong support for overcoming the bottleneck of phytochemical identification. For a predicted formula and experimental MS/MS spectrum, the MS-FINDER program could sort the candidate compounds in the public database based on a comprehensive weighted score, and we took the first 20 reliable compounds to seek the target compound in an in-house database. Certainly, those fragmentation pathways could also be deduced and described as Retro-Diels-Alder (RDA) fragmentation reaction, losses of C4H8, C5H8, CH3, CO, CO2 and others. Accordingly, 63 flavonoids were identified, and their in-silico bioactivity were clearly disclosed by some bioinformatics tools. In this experiment, the flavonoids obtained by the four extraction processes were tested by LC-qTOF-MS. We looked for possible Q-markers from these data matrices and then quantified them; their similarities/differences were also described. The results also indicated that the Macroporous Adsorption Resins (MARs) purification is a low cost, environmentally friendly and effective approach.

Chemometrics-enhanced one-dimensional/comprehensive two-dimensional gas chromatographic analysis for bioactive terpenoids and phthalides in Chaihu Shugan San essential oils.

Chemometrics-enhanced one-dimensional/comprehensive two-dimensional gas chromatographic (GC/GC×GC) technologies, were used to explore the compositions of Chaihu Shugan San essential oils, that were extracted from the herbal formulae by different schemes. We have shown that chemometric resolution using gas chromatographic- mass spectrometry (GC-MS) could be used for the qualitative and quantitative analysis of the majority of Terpenoids or Phthalides from herb formulae and single herbs. A GC×GC system was further optimized to achieve the increased peak capacity and the enhanced signal of the hydro-distillation sample (CSSh). When hardware bottleneck resulted from very complex sample, chemometric tools were once again applied to recover the stained information in the second dimension (2D) matrix data. Heuristic evolving latent projections (HELP) could be used for two dimensional (2D) sub-matrixes Xi at n spectral detection channels, after three dimensional (3D) data splitting. For a real 3D data matrix, alternating trilinear decomposition (ATLD) algorithm could conduct regularization for an iterative trilinear decomposition procedure, by Moore-Penrose pseudoinverse computations based on singular value decomposition. After retention indices (RI) confirmation, 216 target analytes (terpenoids or phthalides) could be elucidated both in CSSh and in supercritical fluid extract (CSSs). Based on the obtained data, some potential quality markers (Q-markers) were identified which may affect the quality of the products. Finally, a "connectivity map" was plotted to describe the unique mechanisms of tradition Chinese medicine (TCM).