‘Ring Breaker’: Neural Network Driven Synthesis Prediction of the Ring System Chemical Space

Ring systems in pharmaceuticals, agrochemicals and dyes are ubiquitous chemical motifs. Whilst the synthesis of common ring systems is well described, and novel ring systems can be readily computationally enumerated, the synthetic accessibility of unprecedented ring systems remains a challenge. ‘Ring Breaker’ uses a data-driven approach to enable the prediction of ring-forming reactions, for which we have demonstrated its utility on frequently found and unprecedented ring systems, in agreement with literature syntheses. We demonstrate the performance of the neural network on a range of ring fragments from the ZINC and DrugBank databases and highlight its potential for incorporation into computer aided synthesis planning tools. These approaches to ring formation and retrosynthetic disconnection offer opportunities for chemists to explore and select more efficient syntheses/synthetic routes.

Download Full-text

‘Ring Breaker’: Assessing Synthetic Accessibility of the Ring System Chemical Space

10.26434/chemrxiv.9938969.v1 ◽

2019 ◽

Author(s):

Amol Thakkar ◽

Nidhal Selmi ◽

Jean-Louis Reymond ◽

Ola Engkvist ◽

Esben Jannik Bjerrum

Keyword(s):

Chemical Space ◽

Ring System ◽

Ring Systems ◽

Multiple Reactions ◽

Zinc Database ◽

Synthetic Accessibility ◽

Synthesis Planning ◽

Reaction Graphs ◽

Synthetic Routes ◽

The Relationship

Ring systems in pharmaceuticals, agrochemicals and dyes are ubiquitous chemical motifs. Whilst the synthesis of common ring systems is well described, and novel ring systems can be readily computationally enumerated, the synthetic accessibility of unprecedented ring systems remains a challenge. ‘Ring Breaker’ enables the prediction of ring-forming reactions, for which we have demonstrated its utility on frequently found and unprecedented ring systems, in agreement with literature syntheses. We demonstrate its performance on a range of ring fragments from the ZINC database and highlight its potential for incorporation into computer aided synthesis planning tools. Additionally, we generate a multi-label dataset using bipartite reaction graphs on which we train ‘Ring Breaker’ to model the relationship between one ring fragment and the multiple reactions recorded for its synthesis in the dataset; we thereby overcome the single-label approaches previously used. These approaches to ring formation and retrosynthetic disconnection offer opportunities for chemists to explore and select more efficient syntheses/synthetic routes.

Download Full-text

‘Ring Breaker’: Assessing Synthetic Accessibility of the Ring System Chemical Space

10.26434/chemrxiv.9938969.v2 ◽

2019 ◽

Author(s):

Amol Thakkar ◽

Nidhal Selmi ◽

Jean-Louis Reymond ◽

Ola Engkvist ◽

Esben Jannik Bjerrum

Keyword(s):

Chemical Space ◽

Ring System ◽

Ring Systems ◽

Multiple Reactions ◽

Zinc Database ◽

Synthetic Accessibility ◽

Synthesis Planning ◽

Reaction Graphs ◽

Synthetic Routes ◽

The Relationship

Ring systems in pharmaceuticals, agrochemicals and dyes are ubiquitous chemical motifs. Whilst the synthesis of common ring systems is well described, and novel ring systems can be readily computationally enumerated, the synthetic accessibility of unprecedented ring systems remains a challenge. ‘Ring Breaker’ enables the prediction of ring-forming reactions, for which we have demonstrated its utility on frequently found and unprecedented ring systems, in agreement with literature syntheses. We demonstrate its performance on a range of ring fragments from the ZINC database and highlight its potential for incorporation into computer aided synthesis planning tools. Additionally, we generate a multi-label dataset using bipartite reaction graphs on which we train ‘Ring Breaker’ to model the relationship between one ring fragment and the multiple reactions recorded for its synthesis in the dataset; we thereby overcome the single-label approaches previously used. These approaches to ring formation and retrosynthetic disconnection offer opportunities for chemists to explore and select more efficient syntheses/synthetic routes.

Download Full-text

Application of deep neural network reveals novel effects of maternal pre-conception exposure to nicotine on rat pup behavior

10.1101/2020.07.16.206961 ◽

2020 ◽

Author(s):

Reza Torabi ◽

Serena Jenkins ◽

Allonna Harker ◽

Ian Q. Whishaw ◽

Robbin Gibb ◽

...

Keyword(s):

Neural Network ◽

Motor Development ◽

Deep Neural Network ◽

Data Driven ◽

Nicotine Exposure ◽

Movement Initiation ◽

Warm Up ◽

Rat Pups ◽

The Neural Network ◽

Data Driven Approach

We present a deep neural network for data-driven analyses of infant rat behavior in an open field task. The network was applied to study the effect of maternal nicotine exposure prior to conception on offspring motor development. The neural network outperformed human expert designed animal locomotion measures in distinguishing rat pups born to nicotine exposed dams versus control dams. Notably, the network discovered novel movement alterations in posture, movement initiation and a stereotypy in warm-up behavior (the initiation of movement along specific dimensions) that were predictive of nicotine exposure. The results suggest that maternal preconception nicotine exposure delays and alters offspring motor development. In summary, we demonstrated that a deep neural network can automatically assess animal behavior with high accuracy, and that it offers a data-driven approach to investigating pharmacological effects on brain development.

Download Full-text

Contributions towards Data driven Deep Learning methods to predict Steady State Fluid Flow in mechanical Heart Valves

Current Directions in Biomedical Engineering ◽

10.1515/cdbme-2021-2159 ◽

2021 ◽

Vol 7 (2) ◽

pp. 625-628

Author(s):

Jan Oldenburg ◽

Julian Renkewitz ◽

Michael Stiehm ◽

Klaus-Peter Schmitz

Keyword(s):

Neural Network ◽

Deep Learning ◽

Steady State ◽

Flow Field ◽

Heart Valves ◽

Data Driven ◽

High Shear ◽

Blood Flows ◽

The Neural Network ◽

Data Driven Approach

Abstract It is commonly accepted that hemodynamic situation is related with cardiovascular diseases as well as clinical post-procedural outcome. In particular, aortic valve stenosis and insufficiency are associated with high shear flow and increased pressure loss. Furthermore, regurgitation, high shear stress and regions of stagnant blood flow are presumed to have an impact on clinical result. Therefore, flow field assessment to characterize the hemodynamic situation is necessary for device evaluation and further design optimization. In-vitro as well as in-silico fluid mechanics methods can be used to investigate the flow through prostheses. In-silico solutions are based on mathematical equitation’s which need to be solved numerically (Computational Fluid Dynamics - CFD). Fundamentally, the flow is physically described by Navier-Stokes. CFD often requires high computational cost resulting in long computation time. Techniques based on deep-learning are under research to overcome this problem. In this study, we applied a deep-learning strategy to estimate fluid flows during peak systolic steady-state blood flows through mechanical aortic valves with varying opening angles in randomly generated aortic root geometries. We used a data driven approach by running 3,500 two dimensional simulations (CFD). The simulation data serves as training data in a supervised deep learning framework based on convolutional neural networks analogous to the U-net architecture. We were able to successfully train the neural network using the supervised data driven approach. The results showing that it is feasible to use a neural network to estimate physiological flow fields in the vicinity of prosthetic heart valves (Validation error below 0.06), by only giving geometry data (Image) into the Network. The neural network generates flow field prediction in real time, which is more than 2500 times faster compared to CFD simulation. Accordingly, there is tremendous potential in the use of AIbased approaches predicting blood flows through heart valves on the basis of geometry data, especially in applications where fast fluid mechanic predictions are desired.

Download Full-text

Machine-learning computation of distance modulus for local galaxies

Astronomy and Astrophysics ◽

10.1051/0004-6361/201936883 ◽

2020 ◽

Vol 635 ◽

pp. A124

Author(s):

A. A. Elyiv ◽

O. V. Melnyk ◽

I. B. Vavilova ◽

D. V. Dobrycheva ◽

V. E. Karachentseva

Keyword(s):

Neural Network ◽

Machine Learning ◽

Angular Size ◽

Data Driven ◽

Gradient Boosting ◽

Distance Modulus ◽

Explanatory Variables ◽

Photometric Redshift ◽

The Neural Network ◽

Data Driven Approach

Context. Quickly growing computing facilities and an increasing number of extragalactic observations encourage the application of data-driven approaches to uncover hidden relations from astronomical data. In this work we raise the problem of distance reconstruction for a large number of galaxies from available extensive observations. Aims. We propose a new data-driven approach for computing distance moduli for local galaxies based on the machine-learning regression as an alternative to physically oriented methods. We use key observable parameters for a large number of galaxies as input explanatory variables for training: magnitudes in U, B, I, and K bands, corresponding colour indices, surface brightness, angular size, radial velocity, and coordinates. Methods. We performed detailed tests of the five machine-learning regression techniques for inference of m−M: linear, polynomial, k-nearest neighbours, gradient boosting, and artificial neural network regression. As a test set we selected 91 760 galaxies at z < 0.2 from the NASA/IPAC extragalactic database with distance moduli measured by different independent redshift methods. Results. We find that the most effective and precise is the neural network regression model with two hidden layers. The obtained root–mean–square error of 0.35 mag, which corresponds to a relative error of 16%, does not depend on the distance to galaxy and is comparable with methods based on the Tully–Fisher and Fundamental Plane relations. The proposed model shows a 0.44 mag (20%) error in the case of spectroscopic redshift absence and is complementary to existing photometric redshift methodologies. Our approach has great potential for obtaining distance moduli for around 250 000 galaxies at z < 0.2 for which the above-mentioned parameters are already observed.

Download Full-text

Estimation of Site Amplification from Geotechnical Array Data Using Neural Networks

Bulletin of the Seismological Society of America ◽

10.1785/0120200346 ◽

2021 ◽

Author(s):

Daniel Roten ◽

Kim B. Olsen

Keyword(s):

Neural Network ◽

Neural Networks ◽

Site Response ◽

Site Amplification ◽

Data Driven ◽

Vertical Array ◽

The Neural Network ◽

Simplifying Assumptions ◽

The Mean ◽

Fully Connected

ABSTRACT We use deep learning to predict surface-to-borehole Fourier amplification functions (AFs) from discretized shear-wave velocity profiles. Specifically, we train a fully connected neural network and a convolutional neural network using mean AFs observed at ∼600 KiK-net vertical array sites. Compared with predictions based on theoretical SH 1D amplifications, the neural network (NN) results in up to 50% reduction of the mean squared log error between predictions and observations at sites not used for training. In the future, NNs may lead to a purely data-driven prediction of site response that is independent of proxies or simplifying assumptions.

Download Full-text

Chemical Space Mapping for Multicomponent Gas Mixtures

10.26434/chemrxiv.14061047.v1 ◽

2021 ◽

Author(s):

Airat Kotliar-Shapirov ◽

Fedor S. Fedorov ◽

Henni Ouerdane ◽

Stanislav Evlashin ◽

Albert G. Nasibulin ◽

...

Keyword(s):

Cyclic Voltammetry ◽

Data Processing ◽

Electrochemical Sensors ◽

Chemical Space ◽

Chemical Species ◽

Single Species ◽

Data Driven ◽

Multicomponent Gas ◽

Data Driven Approach ◽

Cyclic Voltammetry Data

In our manuscript, we present our protocol for data processing to mitigate the effects of interfering analytes on the identification of the chemical species detected by sensors. Considering NO2 and CO2, we designed electrochemical sensors whose response yielded the cyclic voltammetry data that we analyzed to classify single-species components and their mixtures using a data-driven approach to generate a chemical space where their mixtures can be deconvoluted.

Download Full-text

Data Driven Solutions and Discoveries in Mechanics Using Physics Informed Neural Network

10.20944/preprints202006.0258.v1 ◽

2020 ◽

Author(s):

Qi Zhang ◽

Yilin Chen ◽

Ziyi Yang

Keyword(s):

Neural Network ◽

Differential Equations ◽

Automatic Differentiation ◽

Numerical Solutions ◽

Data Driven ◽

Attractive Alternative ◽

The Finite Element Method ◽

The Neural Network ◽

Engineering Problems ◽

Free Parameters

Deep learning has achieved remarkable success in diverse computer science applications, however, its use in other traditional engineering fields has emerged only recently. In this project, we solved several mechanics problems governed by differential equations, using physics informed neural networks (PINN). The PINN embeds the differential equations into the loss of the neural network using automatic differentiation. We present our developments in the context of solving two main classes of problems: data-driven solutions and data-driven discoveries, and we compare the results with either analytical solutions or numerical solutions using the finite element method. The remarkable achievements of the PINN model shown in this report suggest the bright prospect of the physics-informed surrogate models that are fully differentiable with respect to all input coordinates and free parameters. More broadly, this study shows that PINN provides an attractive alternative to solve traditional engineering problems.

Download Full-text

Bundling in Molecular Dynamics Simulations to Improve Generalization Performance in High-Dimensional Neural Network Potentials

10.26434/chemrxiv.7860032 ◽

2019 ◽

Author(s):

Yasuharu Okamoto

Keyword(s):

Neural Network ◽

Force Field ◽

Data Driven ◽

High Dimensional ◽

Field Calculation ◽

Force Field Calculations ◽

Quantitative Accuracy ◽

Force Field Calculation ◽

Data Driven Approach ◽

Dynamics Simulations

High dimensional neural network potential (HDNNP) is interested as an alternative to classical force field calculations by data-driven approach. HDNNP has an advantage over classical force field calculation, such as being able to handle chemical reactions, but there are many points yet to be understood with respect to the chemical transferability in particular for non-organic compounds. In this paper, we focused on Au13+ and Au11+ clusters and showed that the energy of clusters of different sizes can be predicted by HDNNP with semi-quantitative accuracy.

Download Full-text