scholarly journals Benchmarking the performance of Bayesian optimization across multiple experimental materials science domains

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Qiaohao Liang ◽  
Aldair E. Gongora ◽  
Zekun Ren ◽  
Armi Tiihonen ◽  
Zhe Liu ◽  
...  
Keyword(s):  
Random Forest ◽  
High Throughput ◽  
Surrogate Model ◽  
Time Complexity ◽  
Materials Science ◽  
Bayesian Optimization ◽  
Comparable Performance ◽  
Hyperparameter Selection ◽  
High Throughput Experiments ◽  
Anisotropic Kernels

AbstractBayesian optimization (BO) has been leveraged for guiding autonomous and high-throughput experiments in materials science. However, few have evaluated the efficiency of BO across a broad range of experimental materials domains. In this work, we quantify the performance of BO with a collection of surrogate model and acquisition function pairs across five diverse experimental materials systems. By defining acceleration and enhancement metrics for materials optimization objectives, we find that surrogate models such as Gaussian Process (GP) with anisotropic kernels and Random Forest (RF) have comparable performance in BO, and both outperform the commonly used GP with isotropic kernels. GP with anisotropic kernels has demonstrated the most robustness, yet RF is a close alternative and warrants more consideration because it is free from distribution assumptions, has smaller time complexity, and requires less effort in initial hyperparameter selection. We also raise awareness about the benefits of using GP with anisotropic kernels in future materials optimization campaigns.

scholarly journals Two-step machine learning enables optimized nanoparticle synthesis

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Flore Mekki-Berrada ◽  
Zekun Ren ◽  
Tan Huang ◽  
Wai Kuan Wong ◽  
Fang Zheng ◽  
...  
Keyword(s):  
Machine Learning ◽  
Optical Properties ◽  
Materials Science ◽  
Nanoparticle Synthesis ◽  
Bayesian Optimization ◽  
Absorbance Spectrum ◽  
Nanomaterial Synthesis ◽  
Target Spectrum ◽  
Colour Palette ◽  
Algorithmic Framework

AbstractIn materials science, the discovery of recipes that yield nanomaterials with defined optical properties is costly and time-consuming. In this study, we present a two-step framework for a machine learning-driven high-throughput microfluidic platform to rapidly produce silver nanoparticles with the desired absorbance spectrum. Combining a Gaussian process-based Bayesian optimization (BO) with a deep neural network (DNN), the algorithmic framework is able to converge towards the target spectrum after sampling 120 conditions. Once the dataset is large enough to train the DNN with sufficient accuracy in the region of the target spectrum, the DNN is used to predict the colour palette accessible with the reaction synthesis. While remaining interpretable by humans, the proposed framework efficiently optimizes the nanomaterial synthesis and can extract fundamental knowledge of the relationship between chemical composition and optical properties, such as the role of each reactant on the shape and amplitude of the absorbance spectrum.

scholarly journals Discovering collectively informative descriptors from high-throughput experiments

2009 ◽  
Vol 10 (1) ◽  
Author(s):  
Clark D Jeffries ◽  
William O Ward ◽  
Diana O Perkins ◽  
Fred A Wright
Keyword(s):  
High Throughput ◽  
High Throughput Experiments

A Flexible Database for Combinatorial and High-Throughput Materials Science

2005 ◽  
Vol 24 (1) ◽  
pp. 22-28 ◽  
Author(s):  
Andreas Frantzen ◽  
Daniel Sanders ◽  
Jens Scheidtmann ◽  
Ulrich Simon ◽  
Wilhelm F. Maier
Keyword(s):  
High Throughput ◽  
Materials Science

scholarly journals Enabling Solar Fuels Technology With High Throughput Experimentation

2014 ◽  
Vol 1654 ◽  
Author(s):  
J. M. Gregoire ◽  
J. A. Haber ◽  
S. Mitrovic ◽  
C. Xiang ◽  
S. Suram ◽  
...  
Keyword(s):  
High Throughput ◽  
Materials Science ◽  
Technology Development ◽  
Research Effort ◽  
Solar Fuels ◽  
Transportation Fuels ◽  
High Throughput Experimentation ◽  
Earth Abundant ◽  
Combinatorial Materials ◽  
Joint Center

ABSTRACTThe High Throughput Experimentation (HTE) project of the Joint Center for Artificial Photosynthesis (JCAP, http://solarfuelshub.org/) performs accelerated discovery of new earth-abundant photoabsorbers and electrocatalysts. Through collaboration within the DOE solar fuels hub and with the broader research community, the new materials will be utilized in devices that efficiently convert solar energy, water and carbon dioxide into transportation fuels. JCAP-HTE builds high-throughput pipelines for the synthesis, screening and characterization of photoelectrochemical materials. In addition to a summary of these pipelines, we will describe several new screening instruments for high throughput (photo-)electrochemical measurements. These instruments are not only optimized for screening against solar fuels requirements, but also provide new tools for the broader combinatorial materials science community. We will also describe the high throughput discovery, follow-on verification, and device implementation of a new quaternary metal oxide catalyst. This rapid technology development from discovery to device implementation is a hallmark of the multi-faceted JCAP research effort.

scholarly journals Characterizing Malignant Melanoma Clinically Resembling Seborrheic Keratosis Using Deep Knowledge Transfer

Cancers ◽  
2021 ◽  
Vol 13 (24) ◽  
pp. 6300
Author(s):  
Panagiota Spyridonos ◽  
George Gaitanis ◽  
Aristidis Likas ◽  
Ioannis Bassukas
Keyword(s):  
Public Access ◽  
Classification Model ◽  
Bayesian Optimization ◽  
Support Vector ◽  
Seborrheic Keratosis ◽  
Convolutional Network ◽  
Data Set ◽  
Computer Based ◽  
Hyperparameter Selection

Malignant melanomas resembling seborrheic keratosis (SK-like MMs) are atypical, challenging to diagnose melanoma cases that carry the risk of delayed diagnosis and inadequate treatment. On the other hand, SK may mimic melanoma, producing a ‘false positive’ with unnecessary lesion excisions. The present study proposes a computer-based approach using dermoscopy images for the characterization of SΚ-like MMs. Dermoscopic images were retrieved from the International Skin Imaging Collaboration archive. Exploiting image embeddings from pretrained convolutional network VGG16, we trained a support vector machine (SVM) classification model on a data set of 667 images. SVM optimal hyperparameter selection was carried out using the Bayesian optimization method. The classifier was tested on an independent data set of 311 images with atypical appearance: MMs had an absence of pigmented network and had an existence of milia-like cysts. SK lacked milia-like cysts and had a pigmented network. Atypical MMs were characterized with a sensitivity and specificity of 78.6% and 84.5%, respectively. The advent of deep learning in image recognition has attracted the interest of computer science towards improved skin lesion diagnosis. Open-source, public access archives of skin images empower further the implementation and validation of computer-based systems that might contribute significantly to complex clinical diagnostic problems such as the characterization of SK-like MMs.

scholarly journals Cross-Platform Bayesian Optimization System for Autonomous Biological Assay Development

2021 ◽  
Vol 26 (6) ◽  
pp. 579-590
Author(s):  
Sam Elder ◽  
Carleen Klumpp-Thomas ◽  
Adam Yasgar ◽  
Jameson Travers ◽  
Shayne Frebert ◽  
...  
Keyword(s):  
High Throughput ◽  
High Throughput Screening ◽  
Assay Development ◽  
Bayesian Optimization ◽  
Biological Assay ◽  
Proof Of Concept ◽  
Screening Assay ◽  
High Throughput Screening Assay ◽  
Cross Platform ◽  
Assay Conditions

Current high-throughput screening assay optimization is often a manual and time-consuming process, even when utilizing design-of-experiment approaches. A cross-platform, Cloud-based Bayesian optimization-based algorithm was developed as part of the National Center for Advancing Translational Sciences (NCATS) ASPIRE (A Specialized Platform for Innovative Research Exploration) Initiative to accelerate preclinical drug discovery. A cell-free assay for papain enzymatic activity was used as proof of concept for biological assay development and system operationalization. Compared with a brute-force approach that sequentially tested all 294 assay conditions to find the global optimum, the Bayesian optimization algorithm could find suitable conditions for optimal assay performance by testing 21 assay conditions on average, with up to 20 conditions being tested simultaneously, as confirmed by repeated simulation. The algorithm could achieve a sevenfold reduction in costs for lab supplies and high-throughput experimentation runtime, all while being controlled from a remote site through a secure connection. Based on this proof of concept, this technology is expected to be applied to more complex biological assays and automated chemistry reaction screening at NCATS, and should be transferable to other institutions. Graphical Abstract

INFERRING PROTEIN-PROTEIN INTERACTIONS FROM MESSENGER RNA EXPRESSION PROFILES WITH SVM

2005 ◽  
Vol 13 (03) ◽  
pp. 287-298 ◽  
Author(s):  
JUN CAI ◽  
YING HUANG ◽  
LIANG JI ◽  
YANDA LI
Keyword(s):  
High Throughput ◽  
Protein Interactions ◽  
Messenger Rna ◽  
Expression Profiles ◽  
Support Vector ◽  
Svm Classifier ◽  
Good Prediction ◽  
Protein Protein Interactions ◽  
Protein Protein Interaction ◽  
High Throughput Experiments

In post-genomic biology, researchers in the field of proteome focus their attention on the networks of protein interactions that control the lives of cells and organisms. Protein-protein interactions play a useful role in dynamic cellular machinery. In this paper, we developed a method to infer protein-protein interactions based on the theory of support vector machine (SVM). For a given pair of proteins, a new strategy of calculating cross-correlation function of mRNA expression profiles was used to encode SVM vectors. We compared the performance with other methods of inferring protein-protein interaction. Results suggested that, through five-fold cross validation, our SVM model achieved a good prediction. It enables us to show that expression profiles in transcription level can be used to distinguish physical or functional interactions of proteins as well as sequence contents. Lastly, we applied our SVM classifier to evaluate data quality of interaction data sets from four high-throughput experiments. The results show that high-throughput experiments sacrifice some accuracy in determination of interactions because of limitation of experiment technologies.

Sign in / Sign up

Export Citation Format

Share Document