ChemML: A machine learning and informatics program package for the analysis, mining, and modeling of chemical and materials data

Mojtaba Haghighatlari; Gaurav Vishwakarma; Doaa Altarawy; Ramachandran Subramanian; Bhargava U. Kota; Aditya Sonpal; Srirangaraj Setlur; Johannes Hachmann

doi:10.1002/wcms.1458

ChemML: A machine learning and informatics program package for the analysis, mining, and modeling of chemical and materials data

Mojtaba Haghighatlari, Gaurav Vishwakarma, Doaa Altarawy, Ramachandran Subramanian, Bhargava U. Kota, Aditya Sonpal, Srirangaraj Setlur, Johannes Hachmann

Research output: Contribution to journal › Article › peer-review

48 Scopus citations

Abstract

ChemML is an open machine learning (ML) and informatics program suite that is designed to support and advance the data-driven research paradigm that is currently emerging in the chemical and materials domain. ChemML allows its users to perform various data science tasks and execute ML workflows that are adapted specifically for the chemical and materials context. Key features are automation, general-purpose utility, versatility, and user-friendliness in order to make the application of modern data science a viable and widely accessible proposition in the broader chemistry and materials community. ChemML is also designed to facilitate methodological innovation, and it is one of the cornerstones of the software ecosystem for data-driven in silico research. This article is categorized under: Software > Simulation Methods Computer and Information Science > Chemoinformatics Structure and Mechanism > Computational Materials Science Software > Molecular Modeling.

Original language	English
Article number	e1458
Journal	Wiley Interdisciplinary Reviews: Computational Molecular Science
Volume	10
Issue number	4
DOIs	https://doi.org/10.1002/wcms.1458
State	Published - Jul 1 2020
Externally published	Yes

Funding

This work is supported by the National Science Foundation (NSF) CAREER program (grant No. OAC‐1751161), and the New York State Center of Excellence in Materials Informatics (grants No. CMI‐1140384 and CMI‐1148092). Early work on was supported by start‐up funds provided through the University at Buffalo (UB). The deep eutectic solvent application study was funded by the Army Armament Research, Development and Engineering Center (ARDEC) SBIR program (grant No. W15QKN‐17‐C‐0078), and solubility parameter work by Toyota Motor Engineering and Manufacturing North America. is interfaced with the Open Chemistry platform and the MaDE@UB toolkit, and these efforts are supported by the Department of Energy SBIR program (grant No. DE‐SC0017193) and the NSF DIBBs program (grant No. OAC‐1640867), respectively. The DIBBs grant also funded the implementation of several methods of particular interest for MaDE@UB into , such as the Magpie library, the meta data parser, and standard DNNs using Keras. Computing time on the high‐performance computing clusters “,” “,” “,” and “” was provided by the UB Center for Computational Research (CCR). The work presented in this paper is a central part of M.H.'s PhD thesis. M.H. gratefully acknowledges support by Phase‐I and Phase‐II Software Fellowships (grant No. ACI‐1547580‐479590) of the NSF Molecular Sciences Software Institute (grant No. ACI‐1547580) at Virginia Tech. We thank the other members—past and present—of the Hachmann group as well as Profs. Venugopal Govindaraju and Krishna Rajan (both UB) for valuable discussions and insights that have helped guide the development of . ChemML ChemML ChemML Rush Alpha Beta Gamma ChemML Armament Research, Development and Engineering Center, Grant/Award Number: W15QKN‐17‐C‐0078; National Science Foundation, Grant/Award Numbers: ACI‐1547580, OAC‐1640867, OAC‐1751161; New York Center of Excellence in Materials Informatics, Grant/Award Numbers: CMI‐1140384, CMI‐1148092; Office of Science, Grant/Award Number: DE‐SC0017193 Funding information This work is supported by the National Science Foundation (NSF) CAREER program (grant No. OAC-1751161), and the New York State Center of Excellence in Materials Informatics (grants No. CMI-1140384 and CMI-1148092). Early work on ChemML was supported by start-up funds provided through the University at Buffalo (UB). The deep eutectic solvent application study was funded by the Army Armament Research, Development and Engineering Center (ARDEC) SBIR program (grant No. W15QKN-17-C-0078), and solubility parameter work by Toyota Motor Engineering and Manufacturing North America. ChemML is interfaced with the Open Chemistry platform and the MaDE@UB toolkit, and these efforts are supported by the Department of Energy SBIR program (grant No. DE-SC0017193) and the NSF DIBBs program (grant No. OAC-1640867), respectively. The DIBBs grant also funded the implementation of several methods of particular interest for MaDE@UB into ChemML, such as the Magpie library, the meta data parser, and standard DNNs using Keras. Computing time on the high-performance computing clusters ?Rush,? ?Alpha,? ?Beta,? and ?Gamma? was provided by the UB Center for Computational Research (CCR). The work presented in this paper is a central part of M.H.'s PhD thesis. M.H. gratefully acknowledges support by Phase-I and Phase-II Software Fellowships (grant No. ACI-1547580-479590) of the NSF Molecular Sciences Software Institute (grant No. ACI-1547580) at Virginia Tech. We thank the other members?past and present?of the Hachmann group as well as Profs. Venugopal Govindaraju and Krishna Rajan (both UB) for valuable discussions and insights that have helped guide the development of ChemML.

Funders	Funder number
Department of Energy SBIR	ACI‐1547580‐479590, DE‐SC0017193
NSF Molecular Sciences Software Institute
New York Center of Excellence in Materials Informatics
New York State Center of Excellence in Materials Informatics	CMI‐1140384, CMI‐1148092
UB Center for Computational Research
National Science Foundation	1640867, OAC‐1640867, OAC‐1751161, 1751161, ACI‐1547580
Office of Science
University at Buffalo
Armament Research, Development and Engineering Center	W15QKN‐17‐C‐0078
Toyota Motor Engineering and Manufacturing North America

Keywords

data science
data-driven research
informatics
machine learning
program package

Access to Document

10.1002/wcms.1458

Cite this

Haghighatlari, M., Vishwakarma, G., Altarawy, D., Subramanian, R., Kota, B. U., Sonpal, A., Setlur, S., & Hachmann, J. (2020). ChemML: A machine learning and informatics program package for the analysis, mining, and modeling of chemical and materials data. Wiley Interdisciplinary Reviews: Computational Molecular Science, 10(4), Article e1458. https://doi.org/10.1002/wcms.1458

@article{56c5be756b4741fcb3ab6f6ddf637ce1,

title = "ChemML: A machine learning and informatics program package for the analysis, mining, and modeling of chemical and materials data",

abstract = "ChemML is an open machine learning (ML) and informatics program suite that is designed to support and advance the data-driven research paradigm that is currently emerging in the chemical and materials domain. ChemML allows its users to perform various data science tasks and execute ML workflows that are adapted specifically for the chemical and materials context. Key features are automation, general-purpose utility, versatility, and user-friendliness in order to make the application of modern data science a viable and widely accessible proposition in the broader chemistry and materials community. ChemML is also designed to facilitate methodological innovation, and it is one of the cornerstones of the software ecosystem for data-driven in silico research. This article is categorized under: Software > Simulation Methods Computer and Information Science > Chemoinformatics Structure and Mechanism > Computational Materials Science Software > Molecular Modeling.",

keywords = "data science, data-driven research, informatics, machine learning, program package",

author = "Mojtaba Haghighatlari and Gaurav Vishwakarma and Doaa Altarawy and Ramachandran Subramanian and Kota, {Bhargava U.} and Aditya Sonpal and Srirangaraj Setlur and Johannes Hachmann",

note = "Publisher Copyright: {\textcopyright} 2020 Wiley Periodicals, Inc.",

year = "2020",

month = jul,

day = "1",

doi = "10.1002/wcms.1458",

language = "English",

volume = "10",

journal = "Wiley Interdisciplinary Reviews: Computational Molecular Science",

issn = "1759-0876",

publisher = "wiley",

number = "4",

}

ChemML: A machine learning and informatics program package for the analysis, mining, and modeling of chemical and materials data. / Haghighatlari, Mojtaba; Vishwakarma, Gaurav; Altarawy, Doaa et al.
In: Wiley Interdisciplinary Reviews: Computational Molecular Science, Vol. 10, No. 4, e1458, 01.07.2020.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - ChemML

T2 - A machine learning and informatics program package for the analysis, mining, and modeling of chemical and materials data

AU - Haghighatlari, Mojtaba

AU - Vishwakarma, Gaurav

AU - Altarawy, Doaa

AU - Subramanian, Ramachandran

AU - Kota, Bhargava U.

AU - Sonpal, Aditya

AU - Setlur, Srirangaraj

AU - Hachmann, Johannes

PY - 2020/7/1

Y1 - 2020/7/1

N2 - ChemML is an open machine learning (ML) and informatics program suite that is designed to support and advance the data-driven research paradigm that is currently emerging in the chemical and materials domain. ChemML allows its users to perform various data science tasks and execute ML workflows that are adapted specifically for the chemical and materials context. Key features are automation, general-purpose utility, versatility, and user-friendliness in order to make the application of modern data science a viable and widely accessible proposition in the broader chemistry and materials community. ChemML is also designed to facilitate methodological innovation, and it is one of the cornerstones of the software ecosystem for data-driven in silico research. This article is categorized under: Software > Simulation Methods Computer and Information Science > Chemoinformatics Structure and Mechanism > Computational Materials Science Software > Molecular Modeling.

AB - ChemML is an open machine learning (ML) and informatics program suite that is designed to support and advance the data-driven research paradigm that is currently emerging in the chemical and materials domain. ChemML allows its users to perform various data science tasks and execute ML workflows that are adapted specifically for the chemical and materials context. Key features are automation, general-purpose utility, versatility, and user-friendliness in order to make the application of modern data science a viable and widely accessible proposition in the broader chemistry and materials community. ChemML is also designed to facilitate methodological innovation, and it is one of the cornerstones of the software ecosystem for data-driven in silico research. This article is categorized under: Software > Simulation Methods Computer and Information Science > Chemoinformatics Structure and Mechanism > Computational Materials Science Software > Molecular Modeling.

KW - data science

KW - data-driven research

KW - informatics

KW - machine learning

KW - program package

UR - http://www.scopus.com/inward/record.url?scp=85078942145&partnerID=8YFLogxK

U2 - 10.1002/wcms.1458

DO - 10.1002/wcms.1458

M3 - Article

AN - SCOPUS:85078942145

SN - 1759-0876

VL - 10

JO - Wiley Interdisciplinary Reviews: Computational Molecular Science

JF - Wiley Interdisciplinary Reviews: Computational Molecular Science

IS - 4

M1 - e1458

ER -

ChemML: A machine learning and informatics program package for the analysis, mining, and modeling of chemical and materials data

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this