scholarly journals Transformation Rule-Based Molecular Evolution for Automatic Gasoline Molecule Design

2021 ◽  
Author(s):  
Guangqing Cai ◽  
zhefu Liu ◽  
Linzhou Zhang
Keyword(s):  
Molecular Evolution ◽  
Molecular Design ◽  
Chemical Space ◽  
Transformation Rule ◽  
Synthesis Rate ◽  
Initial Structure ◽  
Aviation Gasoline ◽  
Rule Based ◽  
Molecule Design

Automatic molecular design on computers is an emerging technology for the determination of optimal fuel molecules. We developed a computer-aided molecular design framework through a transformation rule-based molecular evolution method. The reaction rule was used as the elementary step to change the molecular structure. A molecule can achieve structural variation continuously using a series of reaction rules. The finding of the optimal molecule can be seen as the evolution of structure in the chemical space, which was guided by using a global optimization algorithm to select the best reaction routine. We showed that the optimized molecule is independent of the input initial structure, proving the robustness of the method. We then applied the method to design gasoline molecules for motor and aviation gasoline. The designed molecules can not only serve as competitive candidate components for high-quality gasoline, but also accelerate the synthesis rate of new molecules in the laboratory.

scholarly journals Multi-fidelity prediction of molecular optical peaks with deep learning

2021 ◽  
Author(s):  
Kevin Greenman ◽  
William Green ◽  
Rafael Gómez-Bombarelli
Keyword(s):  
Optical Properties ◽  
Statistical Methods ◽  
Message Passing ◽  
Density Functional ◽  
Molecular Design ◽  
Chemical Space ◽  
Epistemic Uncertainty ◽  
Computational Cost ◽  
Regression Tree ◽  
Td Dft

Optical properties are central to molecular design for many applications, including solar cells and biomedical imaging. A variety of ab initio and statistical methods have been developed for their prediction, each with a trade-off between accuracy, generality, and cost. Existing theoretical methods such as time-dependent density functional theory (TD-DFT) are generalizable across chemical space because of their robust physics-based foundations but still exhibit random and systematic errors with respect to experiment despite their high computational cost. Statistical methods can achieve high accuracy at a lower cost, but data sparsity and unoptimized molecule and solvent representations often limit their ability to generalize. Here, we utilize directed message passing neural networks (D-MPNNs) to represent both dye molecules and solvents for predictions of molecular absorption peaks in solution. Additionally, we demonstrate a multi-fidelity approach based on an auxiliary model trained on over 28,000 TD-DFT calculations that further improves accuracy and generalizability, as shown through rigorous splitting strategies. Combining several openly-available experimental datasets, we benchmark these methods against a state-of-the-art regression tree algorithm and compare the D-MPNN solvent representation to several alternatives. Finally, we explore the interpretability of the learned representations using dimensionality reduction and evaluate the use of ensemble variance as an estimator of the epistemic uncertainty in our predictions of molecular peak absorption in solution. The prediction methods proposed herein can be integrated with active learning, generative modeling, and experimental workflows to enable the more rapid design of molecules with targeted optical properties.

scholarly journals De Novo Molecule Design Through Molecular Generative Model Conditioned by 3D Information of Protein Binding Sites

2020 ◽  
Author(s):  
Mingyuan Xu ◽  
Ting Ran ◽  
Hongming Chen
Keyword(s):  
Protein Binding ◽  
De Novo ◽  
Structural Information ◽  
Chemical Space ◽  
Three Dimensional ◽  
Binding Pocket ◽  
Generative Model ◽  
Coarse Grained ◽  
Molecule Design ◽  
3D Information

<p><i>De novo</i> molecule design through molecular generative model is gaining increasing attention in recent years. Here a novel generative model was proposed by integrating the 3D structural information of the protein binding pocket into the conditional RNN (cRNN) model to control the generation of drug-like molecules. In this model, the composition of protein binding pocket is effectively characterized through a coarse-grain strategy and the three-dimensional information of the pocket can be represented by the sorted eigenvalues of the coulomb matrix (EGCM) of the coarse-grained atoms composing the binding pocket. In current work, we used our EGCM method and a previously reported binding pocket descriptor DeeplyTough to train cRNN models and compared their performance. It has been shown that the molecules generated with the control of protein environment information have a clear tendency on generating compounds with higher similarity to the original X-ray bound ligand than normal RNN model and also achieving better performance in terms of docking scores. Our results demonstrate the potential application of EGCM controlled generative model for the targeted molecule generation and guided exploration on the drug-like chemical space. </p><p> </p>

scholarly journals Comparison Between SMILES-Based Differential Neural Computer and Recurrent Neural Network Architectures for De Novo Molecule Design

2019 ◽  
Author(s):  
Simon Johansson ◽  
Oleksii Ptykhodko ◽  
Josep Arús-Pous ◽  
Ola Engkvist ◽  
Hongming Chen
Keyword(s):  
De Novo ◽  
Molecular Design ◽  
Research Area ◽  
Computational Time ◽  
Test Cases ◽  
Sequential Data ◽  
Network Architectures ◽  
De Novo Molecular Design ◽  
Molecule Design ◽  
The Cost

In recent years, deep learning for de novo molecular generation has become a rapidly growing research area. Recurrent neural networks (RNN) using the SMILES molecular representation is one of the most common approaches used. Recent study shows that the differentiable neural computer (DNC) can make considerable improvement over the RNN for modeling of sequential data. In the current study, DNC has been implemented as an extension to REINVENT, an RNN-based model that has already been used successfully to make de novo molecular design. The model was benchmarked on its capacity to learn the SMILES language on the GDB-13 and MOSES datasets. The DNC shows improvement on all test cases conducted at the cost of significantly increased computational time and memory consumption.

On Sharp Boundary Problem in Rule Based Expert Systems in the Medical Domain

2010 ◽  
Vol 5 (3) ◽  
pp. 14-26
Author(s):  
Oladipupo O. Olufunke ◽  
Uwadia O. Charles ◽  
Ayo K. Charles
Keyword(s):  
Coronary Heart Disease ◽  
Heart Disease ◽  
Expert System ◽  
Boundary Problem ◽  
System Approach ◽  
Disease Risk ◽  
Fuzzy Expert System ◽  
Sharp Boundary ◽  
Rule Based

Recently, the application of the conventional rule based expert system for disease risk determination in medical domains has increased. However, a major limitation to the effectiveness of the rule based expert system approach is the sharp boundary problem that leads to underestimation or overestimation of boundary cases, which ultimately affects the accuracy of their recommendation. In this paper, an expert driven approach is used to investigate the viability of a fuzzy expert system in the determination of risk associated with coronary heart disease with regards to the sharp boundary problem in rule based expert system.

scholarly journals Determination of the rockfall source in an urban settlement area by using a rule-based fuzzy evaluation

2006 ◽  
Vol 6 (6) ◽  
pp. 941-954 ◽  
Author(s):  
H. Aksoy ◽  
M. Ercanoglu
Keyword(s):  
Field Studies ◽  
Fuzzy Evaluation ◽  
Hazard Map ◽  
Initiation Mechanism ◽  
Urban Settlement ◽  
Rule Based ◽  
Rockfall Hazard ◽  
Source Areas ◽  
The City

Abstract. The evaluation of the rockfall initiation mechanism and the simulation of the runout behavior is an important issue in the prevention and remedial measures for potential rockfall hazards in highway protection, in forest preservation, and especially in urban settlement areas. In most of the studies in the literature, the extent of the rockfall hazard was determined by various techniques basing on the selection of a rockfall source, generally defined as zones of rock bodies having slope angles higher than a certain value, proposed by general practice. In the present study, it was aimed to carry out a rule-based fuzzy analysis on the discontinuity data of andesites in the city of Ankara, Turkey, in order to bring a different and rather systematic approach to determine the source areas for rockfall hazard in an urban settlement, based on the discontinuity and natural slope features. First, to obtain rock source areas (RSAs), data obtained from the field studies were combined with a rule-based fuzzy evaluation, incorporating the altitude difference, the number of discontinuities, the number of wedges and the number of potential slides as the parameters of the fuzzy sets. After processing the outputs of the rule-based fuzzy system and producing the linguistic definitions, it could be possible to obtain potential RSAs. According to the RSA maps, 1.7% of the study area was found to have "high RSA", and 5.8% of the study area was assigned as "medium RSA". Then, potential rockfall hazard map was prepared. At the final stage, based upon the high and medium RSAs, 3.6% of the study area showed "high rockfall potential", while areal distribution of "medium rockfall potential" was found as 7.9%. Both RSA and potential rockfall hazard map were in accordance with the observations performed in the field.

scholarly journals Isolation of human skeletal muscle myosin heavy chain and actin for measurement of fractional synthesis rates

1998 ◽  
Vol 275 (6) ◽  
pp. E1092-E1099 ◽  
Author(s):  
D. L. Hasten ◽  
G. S. Morris ◽  
S. Ramanadham ◽  
K. E. Yarasheski
Keyword(s):  
Skeletal Muscle ◽  
Myosin Heavy Chain ◽  
Heavy Chain ◽  
Human Skeletal Muscle ◽  
Sodium Dodecyl ◽  
Synthesis Rate ◽  
Simple Method ◽  
Skeletal Muscle Myosin ◽  
Fractional Synthesis

Using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), we have developed a simple method to isolate myosin heavy chain (MHC) and actin from small (60–80 mg) human skeletal muscle samples for the determination of their fractional synthesis rates. The amounts of MHC and actin isolated are adequate for the quantification of [13C]leucine abundance by gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS). Fractional synthesis rates of mixed muscle protein (MMP), MHC, and actin were determined in six healthy young subjects (27 ± 1 yr) after they received a 14-h intravenous infusion (prime = 7.58 μmol/kg body wt, constant infusion = 7.58 μmol ⋅ kg body wt−1 ⋅ h−1) of [1-13C]leucine. The fractional synthesis rates of MMP, MHC, and actin were found to be 0.0468 ± 0.0048, 0.0376 ± 0.0033, and 0.0754 ± 0.0078%/h, respectively. Overall, the synthesis rate of MHC was 20% lower ( P = 0.012), and the synthesis rate of actin was 61% higher ( P = 0.060, not significant) than the MMP synthesis rate. The isolation of these proteins for isotope abundance analysis by GC-C-IRMS provides important information about the synthesis rates of these specific contractile proteins, as opposed to the more general information provided by the determination of MMP synthesis rates.

Sign in / Sign up

Export Citation Format

Share Document