Shape perception in chemistry

Janna Hastings, Colin Batchelor, Mitsuhiro Okada

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Organic chemists make extensive use of a diagrammatic language for designing, exchanging and analysing the features of chemicals. In this language, chemicals are represented on a flat (2D) plane following standard stylistic conventions. In the search for novel drugs and therapeutic agents, vast quantities of chemical data are generated and subjected to virtual screening procedures that harness algorithmic features and complex statistical models. However, in silico approaches do not yet compare to the abilities of experienced chemists in detecting more subtle features relevant for evaluating how likely a molecule is to be suitable to a given purpose. Our hypothesis is that one reason for this discrepancy is that human perceptual capabilities, particularly that of 'gestalt' shape perception, make additional information available to our reasoning processes that are not available to in silico processes. This contribution investigates this hypothesis. Algorithmic and logic-based approaches to representation and automated reasoning with chemical structures are able to efficiently compute certain features, such as detecting presence of specific functional groups. To investigate the specific differences between human and machine capabilities, we focus here on those tasks and chemicals for which humans reliably outperform computers: the detection of the overall shape and parts with specific diagrammatic features, in molecules that are large and composed of relatively homogeneous part types with many cycles. We conduct a study in which we vary the diagrammatic representation from the canonical diagrammatic standard of the chemicals, and evaluate speed of human determination of chemical class. We find that human performance varies with the quality of the pictorial representation, rather than the size of the molecule. This can be contrasted with the fact that machine performance varies with the size of the molecule, and is of course impervious to the quality of diagrammatic representation. This result has implications for the design of hybrid algorithms that take features of the overall diagrammatic aspects of the molecule as input into the feature detection and automated reasoning over chemical structure. It also has the potential to inform the design of interactive systems at the interface between human experts and machines.

Original languageEnglish
Title of host publicationCEUR Workshop Proceedings
PublisherCEUR-WS
Pages83-94
Number of pages12
Volume1007
Publication statusPublished - 2013
Event2nd Interdisciplinary Workshop the Shape of Things - Rio de Janeiro, Brazil
Duration: 2013 Apr 32013 Apr 4

Other

Other2nd Interdisciplinary Workshop the Shape of Things
CountryBrazil
CityRio de Janeiro
Period13/4/313/4/4

Fingerprint

Molecules
Functional groups
Screening

Keywords

  • Cognition
  • Logical reasoning
  • Molecular graph
  • Ontology
  • Shape perception
  • Spatial reasoning

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Hastings, J., Batchelor, C., & Okada, M. (2013). Shape perception in chemistry. In CEUR Workshop Proceedings (Vol. 1007, pp. 83-94). CEUR-WS.

Shape perception in chemistry. / Hastings, Janna; Batchelor, Colin; Okada, Mitsuhiro.

CEUR Workshop Proceedings. Vol. 1007 CEUR-WS, 2013. p. 83-94.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hastings, J, Batchelor, C & Okada, M 2013, Shape perception in chemistry. in CEUR Workshop Proceedings. vol. 1007, CEUR-WS, pp. 83-94, 2nd Interdisciplinary Workshop the Shape of Things, Rio de Janeiro, Brazil, 13/4/3.
Hastings J, Batchelor C, Okada M. Shape perception in chemistry. In CEUR Workshop Proceedings. Vol. 1007. CEUR-WS. 2013. p. 83-94
Hastings, Janna ; Batchelor, Colin ; Okada, Mitsuhiro. / Shape perception in chemistry. CEUR Workshop Proceedings. Vol. 1007 CEUR-WS, 2013. pp. 83-94
@inproceedings{53ec72c88ba64a60906a94386b849057,
title = "Shape perception in chemistry",
abstract = "Organic chemists make extensive use of a diagrammatic language for designing, exchanging and analysing the features of chemicals. In this language, chemicals are represented on a flat (2D) plane following standard stylistic conventions. In the search for novel drugs and therapeutic agents, vast quantities of chemical data are generated and subjected to virtual screening procedures that harness algorithmic features and complex statistical models. However, in silico approaches do not yet compare to the abilities of experienced chemists in detecting more subtle features relevant for evaluating how likely a molecule is to be suitable to a given purpose. Our hypothesis is that one reason for this discrepancy is that human perceptual capabilities, particularly that of 'gestalt' shape perception, make additional information available to our reasoning processes that are not available to in silico processes. This contribution investigates this hypothesis. Algorithmic and logic-based approaches to representation and automated reasoning with chemical structures are able to efficiently compute certain features, such as detecting presence of specific functional groups. To investigate the specific differences between human and machine capabilities, we focus here on those tasks and chemicals for which humans reliably outperform computers: the detection of the overall shape and parts with specific diagrammatic features, in molecules that are large and composed of relatively homogeneous part types with many cycles. We conduct a study in which we vary the diagrammatic representation from the canonical diagrammatic standard of the chemicals, and evaluate speed of human determination of chemical class. We find that human performance varies with the quality of the pictorial representation, rather than the size of the molecule. This can be contrasted with the fact that machine performance varies with the size of the molecule, and is of course impervious to the quality of diagrammatic representation. This result has implications for the design of hybrid algorithms that take features of the overall diagrammatic aspects of the molecule as input into the feature detection and automated reasoning over chemical structure. It also has the potential to inform the design of interactive systems at the interface between human experts and machines.",
keywords = "Cognition, Logical reasoning, Molecular graph, Ontology, Shape perception, Spatial reasoning",
author = "Janna Hastings and Colin Batchelor and Mitsuhiro Okada",
year = "2013",
language = "English",
volume = "1007",
pages = "83--94",
booktitle = "CEUR Workshop Proceedings",
publisher = "CEUR-WS",

}

TY - GEN

T1 - Shape perception in chemistry

AU - Hastings, Janna

AU - Batchelor, Colin

AU - Okada, Mitsuhiro

PY - 2013

Y1 - 2013

N2 - Organic chemists make extensive use of a diagrammatic language for designing, exchanging and analysing the features of chemicals. In this language, chemicals are represented on a flat (2D) plane following standard stylistic conventions. In the search for novel drugs and therapeutic agents, vast quantities of chemical data are generated and subjected to virtual screening procedures that harness algorithmic features and complex statistical models. However, in silico approaches do not yet compare to the abilities of experienced chemists in detecting more subtle features relevant for evaluating how likely a molecule is to be suitable to a given purpose. Our hypothesis is that one reason for this discrepancy is that human perceptual capabilities, particularly that of 'gestalt' shape perception, make additional information available to our reasoning processes that are not available to in silico processes. This contribution investigates this hypothesis. Algorithmic and logic-based approaches to representation and automated reasoning with chemical structures are able to efficiently compute certain features, such as detecting presence of specific functional groups. To investigate the specific differences between human and machine capabilities, we focus here on those tasks and chemicals for which humans reliably outperform computers: the detection of the overall shape and parts with specific diagrammatic features, in molecules that are large and composed of relatively homogeneous part types with many cycles. We conduct a study in which we vary the diagrammatic representation from the canonical diagrammatic standard of the chemicals, and evaluate speed of human determination of chemical class. We find that human performance varies with the quality of the pictorial representation, rather than the size of the molecule. This can be contrasted with the fact that machine performance varies with the size of the molecule, and is of course impervious to the quality of diagrammatic representation. This result has implications for the design of hybrid algorithms that take features of the overall diagrammatic aspects of the molecule as input into the feature detection and automated reasoning over chemical structure. It also has the potential to inform the design of interactive systems at the interface between human experts and machines.

AB - Organic chemists make extensive use of a diagrammatic language for designing, exchanging and analysing the features of chemicals. In this language, chemicals are represented on a flat (2D) plane following standard stylistic conventions. In the search for novel drugs and therapeutic agents, vast quantities of chemical data are generated and subjected to virtual screening procedures that harness algorithmic features and complex statistical models. However, in silico approaches do not yet compare to the abilities of experienced chemists in detecting more subtle features relevant for evaluating how likely a molecule is to be suitable to a given purpose. Our hypothesis is that one reason for this discrepancy is that human perceptual capabilities, particularly that of 'gestalt' shape perception, make additional information available to our reasoning processes that are not available to in silico processes. This contribution investigates this hypothesis. Algorithmic and logic-based approaches to representation and automated reasoning with chemical structures are able to efficiently compute certain features, such as detecting presence of specific functional groups. To investigate the specific differences between human and machine capabilities, we focus here on those tasks and chemicals for which humans reliably outperform computers: the detection of the overall shape and parts with specific diagrammatic features, in molecules that are large and composed of relatively homogeneous part types with many cycles. We conduct a study in which we vary the diagrammatic representation from the canonical diagrammatic standard of the chemicals, and evaluate speed of human determination of chemical class. We find that human performance varies with the quality of the pictorial representation, rather than the size of the molecule. This can be contrasted with the fact that machine performance varies with the size of the molecule, and is of course impervious to the quality of diagrammatic representation. This result has implications for the design of hybrid algorithms that take features of the overall diagrammatic aspects of the molecule as input into the feature detection and automated reasoning over chemical structure. It also has the potential to inform the design of interactive systems at the interface between human experts and machines.

KW - Cognition

KW - Logical reasoning

KW - Molecular graph

KW - Ontology

KW - Shape perception

KW - Spatial reasoning

UR - http://www.scopus.com/inward/record.url?scp=84923684354&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923684354&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84923684354

VL - 1007

SP - 83

EP - 94

BT - CEUR Workshop Proceedings

PB - CEUR-WS

ER -