The discovery of novel materials and functional molecules can help to solve some of society's most urgent challenges, ranging from efficient energy harvesting and storage to uncovering novel ...pharmaceutical drug candidates. Traditionally matter engineering-generally denoted as inverse design-was based massively on human intuition and high-throughput virtual screening. The last few years have seen the emergence of significant interest in computer-inspired designs based on evolutionary or deep learning methods. The major challenge here is that the standard strings molecular representation SMILES shows substantial weaknesses in that task because large fractions of strings do not correspond to valid molecules. Here, we solve this problem at a fundamental level and introduce SELFIES (SELF-referencIng Embedded Strings), a string-based representation of molecules which is 100% robust. Every SELFIES string corresponds to a valid molecule, and SELFIES can represent every molecule. SELFIES can be directly applied in arbitrary machine learning models without the adaptation of the models; each of the generated molecule candidates is valid. In our experiments, the model's internal memory stores two orders of magnitude more diverse molecules than a similar test with SMILES. Furthermore, as all molecules are valid, it allows for explanation and interpretation of the internal working of the generative models.
Display omitted
•Recent developments in deep learning for inverse molecular design are reviewed.•Challenges and opportunities for three directions are identified and discussed.•GNNs are seen as ...transformative catalyst towards developing 3D representations.•Chemical text mining and QC computations can contribute to growing property data.•Progress in generative models and RL is poised to improve current design methods.
The discovery of superior molecular solutions through computational methods is critical for innovative technologies and their role in addressing pressing resources, health, and environmental issues. Despite its short timespan, the synergetic application of deep learning to inverse molecular design has outpaced decades of theoretical efforts, bearing promise to transform current molecular design paradigms. Herein, we provide an overview of the element of computational inverse molecular design and offer our views on current limitations and outstanding challenges. In our perspective, three main directions are identified for each element and analyzed in terms of their merits and relevant novel deep learning developments. For the molecular representations element, Graph Neural Networks (GNNs), grids, and knowledge graphs (KGs) are discussed for enhancing the expressivity, complexity, descriptivity of relevant molecular information, respectively. Second, chemical text mining, accelerated quantum chemical calculations, and transfer learning are explored to augment the size and the accuracy of current property data and predictive models. Last, emerging trends in design methods including generative modeling, reinforcement learning (RL), and active learning (AL) are examined for optimizing not only computational costs, but also experimental and simulation efforts. The presented discussions are aimed at catalyzing progress and interdisciplinary collaborations toward general-purpose inverse design frameworks.
The construction of macromolecular models for the amorphous structure of coal can help reveal its physicochemical properties from a microscopic perspective and provide insight into its reaction ...mechanisms, leading to the development of cleaner coal technologies. However, this process requires careful consideration of characterization information. Researchers often need to intervene manually, which makes the task time-consuming. In this study, we proposed a multi-modal deep learning technique, namely ClipIRMol (contrastive language-image pre-training for infrared-molecule), for predicting coal molecular fragments based on the reverse molecular design method. On this basis, a structure evolution algorithm was developed to transform these fragments into a complex molecular structure model. Our approach takes elemental analysis, IR spectrum, and 13C NMR data as inputs. It is capable of constructing highly accurate molecular models of any different types of coal with atom count ranging from tens to thousands in just a few minutes. These spectra were simulated by quantum chemical calculations to show alignment with their experimental data. The introduced 3D molecular models grounded in topological structures overcome the limitation of traditional nearly-planar structures. This offers a new direction for macromolecular modeling of amorphous organic macromolecules.
Display omitted
•ClipIRMol and structure evolution algorithm quickly constructs coals model on different scales.•Coal evolution algorithm incorporates expert experience to create an interpretable 3D model.•ClipIRMol characterized IR spectra and molecules based on comparative learning model.•Molecular models of coal generated is rigorously validated using DFT.•ClipIRMol's versatility can extend to other organic macromolecules.
Chronic Inflammation is associated with various types of diseases that involves pro-inflammatory cytokines like IL-6 and TNF-α. High costs and serious side effects of available ...anti-inflammatory/immunomodulatory drugs led us to design new compounds with promising anti-inflammatory activities. Many drugs and biologically important compounds involve naphthoquinone and thiazole moieties in their core structures. Thereby, here we report the synthesis, characterization and anti-inflammatory activities of new naphthoquinone thiazole hybrids by reaction of naphthoquinone acyl thioureas with various α-bromoketone derivatives. The position of NO2 group in one of the phenyl rings of naphthoquinone thiazole hybrids was changed while different substituents were introduced at the para position of the second phenyl ring. All compounds were tested for potential immunomodulatory effect. No inflammatory cytokines were observed in the absence of LPS stimulant. On the other hand, they had promising anti-inflammatory immunomodulatory activities by being able to decrease the production of the pro-inflammatory cytokines (TNF-α and IL-6) in the LPS-stimulated cells. In an effort to find the possible mechanism of action, several enzymes involved in signalling pathways that play critical roles in inflammatory responses were screened in silico. Subsequent to inverse molecular docking approach, PI3K was predicted be the potential target. The docked complexes of the most potent compounds 5g and 5i were subjected to molecular dynamics simulation to assess the binding stability of the igands with the putative target. Acid dissociation constants (pKa) of the products were also determined potentiometrically.
Research efforts are placing an ever increasing emphasis on identifying signal transduction pathways related to the chemopreventive activity of curcumin. Its anticarcinogenic effects are presumably ...mediated by the regulation of signaling cascades, including nuclear factor κB (NF-κB), activator protein 1 (AP-1), and mitogen-activated protein kinases (MAPK). By modulating signal transduction pathways, curcumin induces apoptosis in malignant cells, thus inhibiting cancer development and progression. Due to the lack of mechanistic insight in the scientific literature, we developed a novel inverse molecular docking protocol based on the CANDOCK algorithm. For the first time, we performed inverse molecular docking of curcumin into a collection of 13,553 available human protein structures from the Protein Data Bank resulting in prioritized target proteins of curcumin. Our predictions were in agreement with the scientific literature and confirmed that curcumin binds to folate receptor β, DNA (cytosine-5)-methyltransferase 3A, metalloproteinase-2, mitogen-activated protein kinase 9, epidermal growth factor receptor and apoptosis-inducing factor 1. We also identified new potential protein targets of curcumin, namely deoxycytidine kinase, NAD-dependent protein deacetylase sirtuin-1 and -2, ecto-5'-nucleotidase, core histone macro-H2A.1, tyrosine-protein phosphatase non-receptor type 11, macrophage colony-stimulating factor 1 receptor, GTPase HRas, aflatoxin B1 aldehyde reductase member 3, aldo-keto reductase family 1 member C3, amiloride-sensitive amine oxidase, death-associated protein kinase 2 and tryptophan-tRNA ligase, that may all play a crucial role in its observed anticancer effects. Moreover, our inverse docking results showed that curcumin potentially binds also to the proteins cAMP-specific 3',5'-cyclic phosphodiesterase 4D and 17-β-hydroxysteroid dehydrogenase type 10, which provides a new explanation for its efficiency in the treatment of Alzheimer's disease. We firmly believe that our computational results will complement and direct future experimental studies on curcumin's anticancer activity as well as on its therapeutic effects against Alzheimer's disease.
In recent years, the synthesis of monomer sequence-defined polymers has expanded into broad-spectrum applications in biomedical, chemical, and materials science fields. Pursuing the characterization ...and inverse design of these polymer systems requires our fundamental understanding not only at the individual monomer level, but also considering the chain scales, such as polymer configuration, self-assembly, and phase separation. However, our accessibility to this field is still rudimentary due to the limitations of traditional design approaches, the complexity of chemical space along with the burdened cost and time issues that prevent us from unveiling the underlying monomer sequence-structure-property relationships. Fortunately, thanks to the recent advancements in molecular dynamics simulations and machine learning (ML) algorithms, the bottlenecks in the tasks of establishing the structure-function correlation of the polymer chains can be overcome. In this review, we will discuss the applications of the integration between ML techniques and coarse-grained molecular dynamics (CGMD) simulations to solve the current issues in polymer science at the chain level. In particular, we focus on the case studies in three important topics-polymeric configuration characterization, feed-forward property prediction, and inverse design-in which CGMD simulations are leveraged to generate training datasets to develop ML-based surrogate models for specific polymer systems and designs. By doing so, this computational hybridization allows us to well establish the monomer sequence-functional behavior relationship of the polymers as well as guide us toward the best polymer chain candidates for the inverse design in undiscovered chemical space with reasonable computational cost and time. Even though there are still limitations and challenges ahead in this field, we finally conclude that this CGMD/ML integration is very promising, not only in the attempt of bridging the monomeric and macroscopic characterizations of polymer materials, but also enabling further tailored designs for sequence-specific polymers with superior properties in many practical applications.
Thiazolidinediones form drugs that treat insulin resistance in type 2 diabetes mellitus. Troglitazone represents the first drug from this family, which was removed from use by the FDA due to its ...hepatotoxicity. As an alternative, rosiglitazone was developed, but it was under the careful watch of FDA for a long time due to suspicion, that it causes cardiovascular diseases, such as heart failure and stroke. We applied a novel inverse molecular docking protocol to discern the potential protein targets of both drugs. Troglitazone and rosiglitazone were docked into predicted binding sites of >67,000 protein structures from the Protein Data Bank and examined. Several new potential protein targets with successfully docked troglitazone and rosiglitazone were identified. The focus was devoted to human proteins so that existing or new potential side effects could be explained or proposed. Certain targets of troglitazone such as 3-oxo-5-beta-steroid 4-dehydrogenase, neutrophil collagenase, stromelysin-1, and VLCAD were pinpointed, which could explain its hepatoxicity, with additional ones indicating that its application could lead to the treatment/development of cancer. Results for rosiglitazone discerned its interaction with members of the matrix metalloproteinase family, which could lead to cancer and neurodegenerative disorders. The concerning cardiovascular side effects of rosiglitazone could also be explained. We firmly believe that our results deepen the mechanistic understanding of the side effects of both drugs, and potentially with further development and research maybe even help to minimize them. On the other hand, the novel inverse molecular docking protocol on the other hand carries the potential to develop into a standard tool to predict possible cross-interactions of drug candidates potentially leading to adverse side effects.
Roem. & Schult and resin of
(L.) BURM. F. are commonly used in Omani traditional medication against various ailments. Herein, their antiproliferative and antioxidant potential was explored. ...Bioassay-guided fractionation of the methanol extract of both plants led to the isolation of 14 known compounds, viz.,
-
from
and
-
from
Their structures were confirmed by combined spectroscopic techniques including 1D (
H and
C) and 2D (HMBC, HSQC, COSY) nuclear magnetic resonance (NMR), and electrospray ionization-mass spectrometry (ESI-MS). The cytotoxic potential of isolates was tested against the triple-negative breast cancer cell line (MDA-MB-231). Compound
exhibited excellent antiproliferative activity in a range of 31 μM, followed by compounds
-
,
, and
, which depicted IC
values in the range of 35-60 μM, while
,
, and
also demonstrated IC
values >72 μM. Subsequently, in silico target fishing was applied to predict the most potential cellular drug targets of the active compounds, using pharmacophore modeling and inverse molecular docking approach. The extensive in silico analysis suggests that our compounds may target carbonic anhydrase II (CA-II) to exert their anticancer activities. When tested on CA-II, compounds
(IC
= 14.4 µM),
(IC
= 23.3), and
(IC
= 24.4 µM) showed excellent biological activities in vitro. Additionally, the ethyl acetate fraction of both plants showed promising antioxidant activity. Among the isolated compounds,
possesses the highest antioxidant (55 μM) activity followed by
(241 μM). The results indicated that compound
can be a promising candidate for antioxidant drugs, while compound
is a potential candidate for anticancer drugs.
The Front Cover picture shows how information gathered from single‐property optimizations can be exploited to effectively, efficiently, and rationally design novel compounds with multiple desired ...properties, especially when constraints are applicable. More information can be found in the Concept by F. De Vleeschouwer et al. on page 1414 in Issue 10, 2016 (DOI: 10.1002/cphc.201501189).
Theoretical design can be an important aid in the discovery of valuable stable radical systems. In this study, inverse molecular design was applied to thiadiazinyl radicals to maximize two ...fundamental electronic properties in chemistry, namely electrophilicity and nucleophilicity. We discovered that the thiadiazinyl structure can be made moderately nucleophilic or highly electrophilic, depending on the electronic character of its substituents. The most electrophilic thiadiazinyl radical displays the highest electrophilic character encountered so far for uncharged radical systems. The accumulated data relating to the influence of different substituents at the various sites has allowed us to propose thiadiazinyl radicals that are intrinsically stable and as electrophilic or nucleophilic as possible. Although the nucleophilicity optimum fulfills the additional requirement of being stable, a significant drop from highly to merely moderately electrophilic is observed for the new electrophilicity optimum.
Inverse molecular design has revealed that the thiadiazinyl structure can be made moderately nucleophilic or highly electrophilic depending on the electronic character of its substituents. Under the additional condition of being stable, the electrophilicity optimum shows a significant drop from highly to merely moderately electrophilic.