Letter Cite This: J. Chem. Inf. Model. XXXX, XXX, XXX−XXX
pubs.acs.org/jcim
Computational Chemistry: A Rising Tide of Women M. Katharine Holloway*,† and Georgia B. McGaughey*,‡ †
Gfree Bio, LLC, 10601 FM 2222, Suite R, PMB 206, Austin, Texas 78730, United States Vertex Pharmaceuticals, 50 Northern Avenue, Boston, Massachusetts 02210, United States
‡
National Girls Collaborative Project,2 little difference was found between the genders at the K−12 level in terms of enrollment, performance, or scores in STEM-related subjects, with the exception of computer science and engineering, where the ratio is slanted toward a higher male class enrollment. After the formative years, however, there is a steady decline as fewer women graduate college in STEM fields, leading to a less diverse pool of potential employees from which to recruit and a marked reduction of women employed in STEM. What happens? There are multiple hypotheses and active research to understand the patterns that lead to this discrepancy. Melinda Gates, for example, has been highly vocal on this topic. In her field of computer science, the representation of women has fallen from a high of 37% in 1984 to 18% currently.3 However, the focus of this letter is not to explore the reasons why there are such disparities, mainly because there are numerous studies on this topic already.4−6 Rather, we choose to focus on how this plays out in computational chemistry, a field in which the coauthors have over 60 years of combined experience and more than 200 publications. Like computational biology, computational chemistry is a relatively young field that has grown symbiotically with exponentially increasing computer power during the “digital age”; like computational biology, it is also a synergistic bridge between two other disciplines, i.e., computer science and chemistry.
ABSTRACT: The authors were inspired to explore the topic of gender diversity in computational chemistry on the basis of similar recent publications in the related fields of medicinal chemistry (Huryn, D. M.; et al. ACS Med. Chem. Lett. 2017, 8, 900) and computational biology (Bonham, K. S.; Stefan, M. I. PLoS Comput. Biol. 2017, 13, e1005134). To do so, we examined historical demographics in two different professional settings, i.e., attendance/participation at the Gordon Research Conferences on Computer-Aided Drug Design and Computational Chemistry and membership in the Computers in Chemistry Division of the American Chemical Society. We conclude that female representation in computational chemistry has risen steadily over the last 40 years and likely stands at around 25%, which appears to slightly exceed that of the neighboring fields of computer science and medicinal chemistry. In accordance with the old slogan that “a rising tide lifts all boats”, here a rising tide of women scientists is having an impact on the field of computational chemistry. Tactics to ensure that this number continues to improve are highlighted.
■
■
DISCUSSION In a public sense, computational chemistry has only fairly recently been recognized as a distinct field of study. Three computational chemists, Paul Crutzen, Mario Molina, and F. Sherwood Rowland, won the 1995 Nobel Prize in Chemistry for constructing mathematical models using thermodynamic and chemical laws. However, modern computational chemistry was not really recognized until 1998 when Walter Kohn and John Pople won the Nobel Prize in Chemistry for their work on density functional theory and computational methods in quantum chemistry. The seminal work of pioneering computational chemists was highlighted again when the 2013 Nobel Prize in Chemistry was awarded to Martin Karplus, Michael Levitt, and Arieh Warshel, who were acknowledged for developing classical and quantum mechanical methods and programs in the 1970s that laid the foundation for many of today’s tools. Despite this recognition, most current scientists in the field did not obtain advanced degrees in computational chemistry but rather followed the track of organic, physical, or medicinal chemistry with a focus on computational work, either in a Ph.D. program or in a subsequent postdoctoral position. As a result, we could not rely on counting the number of degrees awarded in
INTRODUCTION
Gender imbalance is still a significant problem in today’s workforce, and there are numerous reports concerning the paucity of women in political roles, on boards, and in the C-suites of companies. In the area of Science, Technology, Engineering, and Mathematics (STEM), women and men are equally represented going into college (at least in the United States), and then representation becomes imbalanced as educations and careers advance.1 It is important to understand the root cause of this disparity as individuals traverse academia and move onward toward employment. According to the UNESCO Institute for Statistics (UIS), fewer than 30% of STEM researchers worldwide, across all disciplines, are female.1 In a recent study by The © XXXX American Chemical Society
Published: April 6, 2018 A
DOI: 10.1021/acs.jcim.8b00170 J. Chem. Inf. Model. XXXX, XXX, XXX−XXX
Letter
Journal of Chemical Information and Modeling
Figure 1. Comparison of percentages of male and female conferees at QSAR/CADD GRCs (1989−2017).
Figure 2. Comparison of percentages of male and female conferees at Computational Chemistry GRCs (1990−2016).
ACS members have always had the option to select Male, Female, or neither (i.e., gender N/A). An inspection of the number of female attendees at both the QSAR/CADD (Figure 1) and COMP (Figure 2) GRCs shows significant upward trends since 1989 and 1990, respectively. By 1989, the number of female attendees at the QSAR/CADD GRC had jumped to 13%, compared with only 1% in 1975. We believe that some of this change may be attributable to the visibility, influence, and mentorship of one of the female attendees at the 1975 meeting, Yvonne C. Martin, who was invited back to chair the 1977 meeting. As one may imagine, there were not very many female GRC chairs in 1977. The number of women at the QSAR/CADD GRC has continued to grow steadily, peaking at 38% in 2017 (Figure 1). Likewise, the percentage of women attending the COMP GRC has seen steady growth, peaking at 32% in 2016 (Figure 2). Given the somewhat focused nature of these two conferences, we turned our attention to a more diverse, and perhaps more representative, set of computational chemists, i.e., the members of the COMP Division of the ACS. Figure 3 shows demographic data for the COMP Division from 1995 to 2016. Here the number of members identifying as Female peaked at around 18% in 2000 and remained steady for the next decade. However, as noted previously, members are allowed to opt out of gender selfidentification. The green line corresponds to those not selecting Male or Female. This number increased significantly over the 5
computational chemistry in order to explore current and historic gender representation. As an alternate approach, we examined the gender demographics of two professional organizations that include computational chemists as a subgroup, i.e., the American Chemical Society (ACS), which has a Division of Computers in Chemistry (COMP), and the Gordon Research Conferences (GRC), which has sponsored, in alternating years, ComputerAided Drug Design (CADD)7 and Computational Chemistry (COMP) conferences since 1975 and 1986, respectively. We thought that these data might be able to speak to overall gender representation in the field. As with all retrospective data, there are gaps and inconsistencies, some of which illustrate the changing times. For example, despite the fact that applicants were required to state their gender when applying to a GRC to determine shared room assignments, gender statistics were apparently not included in the GRC annual reports (available in the archives at the Chemical Heritage Foundation) until 1990. Prior to that, only the number of “conferees” and “wives and guests” were tabulated! However, we do know that the very first QSAR/ CADD GRC meeting in 1975 had only five female participants out of 94 registered attendees, as determined by inspection of the GRC conference photo and a cross-check with one of the women present. Likewise, there are gaps in the ACS COMP Division demographic records, with no data available prior to 1995. In addition, unlike the GRC, where gender is required for housing, B
DOI: 10.1021/acs.jcim.8b00170 J. Chem. Inf. Model. XXXX, XXX, XXX−XXX
Letter
Journal of Chemical Information and Modeling
Figure 3. Comparison of percentages of male and female members of the ACS COMP Division (1995−2016).
Figure 4. Comparison of percentages of male and female officers of the ACS COMP Division (1994−2017).
However, representation at meetings is not enough. The question remains whether female scientists are involved solely as participants or if they are also being invited to present their research and organize sessions. To address this point, we turned again to the QSAR/CADD GRC data. Figure 5 shows a comparison of percentages of the female conferees, female speakers, and female session chairs. The parity to overrepresentation of female session chairs throughout much of the time range should be noted. Unfortunately, in several years female speakers significantly lagged their representation at the conference (e.g., in 2011). However, the data are clearly encouraging. Overall, there appears to be a rising tide of women attending, contributing to, and deciding the content of professional meetings in the field of computational chemistry. That said, the following strategies are suggested in order to continue to improve the number and participation of women in the field:
year period from 2010 to 2015, with over 30% of members opting out of gender selection in 2015. This makes it hard to determine the current number of female members of the COMP Division with 17% selecting female in 2016. Conservatively, one might estimate that at least a representative number of members without gender identification (i.e., 17%) are likely female. This would add 3.7% to the 17%, bringing the female representation to over 20%. A more generous estimate might be that a large percentage of those not self-identifying are female. The percentage of female members then could be as low as 17% or as high as 39% but is likely somewhere in between. Our best guess might be somewhere around 25%. This estimated demographic compares favorably with that of the Medicinal Chemistry (MEDI) Division of the ACS (20%).8 However, what is more interesting in the COMP Division data is to examine the representation of female members among the elected officers (Figure 4). With the exception of 2015 and 2016, females have occupied 20% or more of the officer positions since 2000, including division chair. In fact, there have been four female COMP chairs in the last 10 years.
1. Educational Awareness: • Ensure that academic faculty reflect a favorable gender balance.9 C
DOI: 10.1021/acs.jcim.8b00170 J. Chem. Inf. Model. XXXX, XXX, XXX−XXX
Letter
Journal of Chemical Information and Modeling
Figure 5. Participation by female conferees, speakers, and session chairs at QSAR/CADD GRCs (1989−2017).
■
• Educate all employees on the topic of unconscious bias,10 e.g., via Diversity and Inclusion courses or through Employee Research Networks (ERNs).11 • Engage women and men in the topic of gender disparity. There are numerous resources to help, e.g., Men Advocating Real Change (MARC).12 • Broadly communicate success stories of women in science. 2. Nominations: • Elect women and other under-represented minorities to the boards of scientific conferences. • Nominate women and minority representatives for scientific awards as well as for service on award selection committees. • Recommend women and minority representatives for speaker lists. 3. Data Analytics: • Track demographics via robust data analytics to measure progress or regression. • Openly publish data trends and patterns.
computational biology was intermediate between the neighboring fields of computer science and biology. Although there are obvious weaknesses in their study (e.g., the authors relied on Gender-API.com to distinguish between female and male names, something which is difficult to do with certain ethnicities), computational chemistry appears comparable in the sense that the percentage of women is greater than in computer science. In contrast to computational biology, however, where the percentage of women in computational biology is less than the percentage of women in biology, the percentage of women in computational chemistry is actually higher than, or at least equal to, that in medicinal chemistry. Vigilance is necessary to ensure that the numbers of women and other under-represented minorities11 continue to rise within the field of computational chemistry. In accordance with the old slogan that “a rising tide lifts all boats”,15 this rising tide of women can help to raise all of the science in our field, as it has been doing since Yvonne C. Martin first stepped forward to chair that 1977 GRC. She recently received the ACS Award for Computers in Chemical and Pharmaceutical Research to recognize her pioneering efforts in the field of computational chemistry. We have also played a part in this rising tide: one of us (M.K.H.) chaired the 2001 CADD GRC and served as the 2011 chair of the ACS COMP Division, and the other (G.B.M.) will chair the 2021 CADD GRC.
CONCLUSION On the basis of an analysis of demographic data for two major computational Gordon Research Conferences and the COMP Division of the ACS, it appears that women are at least as well represented in computational chemistry as in some other chemistry disciplines (e.g., medicinal chemistry) and better represented than in computer science. This may possibly demonstrate that while the gender balance has not reached parity, the number of women is actually higher when the two fields are combined. This is particularly encouraging since computer science is notoriously poor (second only to engineering) as a field of choice for women and has shown a steady decline over the past few decades. In addition, a recent Chemical & Engineering News article13 addressed the general issue of gender imbalance in the drug industry, interviewing a number of female executives, and many of the findings reported there are consistent with the ones presented herein. So, how does gender representation in computational chemistry compare to computational biology, another interdisciplinary field that combines computer science with biology? A recent study14 also showed that female representation in
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. *E-mail:
[email protected]. ORCID
Georgia B. McGaughey: 0000-0002-3586-965X Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
We thank the American Chemical Society, Gordon Research Conferences, and the Chemical Heritage Foundation for making data available for this analysis. We also thank the women (and especially the men) who have contributed to this rising tide. D
DOI: 10.1021/acs.jcim.8b00170 J. Chem. Inf. Model. XXXX, XXX, XXX−XXX
Letter
Journal of Chemical Information and Modeling
■
REFERENCES
(1) UNESCO, Gender and Science. Women in Science - Explore the Data. http://www.unesco.org/new/en/natural-sciences/priority-areas/ gender-and-science/improving-measurement-of-gender-equality-instem/women-in-science-explore-the-data (accessed March 25, 2018). (2) National Girls Collaborative Project: Statistics: State of Girls and Women in STEM. https://ngcproject.org/statistics (accessed March 26, 2018). (3) White, G. B. Melinda Gates: The Tech Industry Needs to Fix Its Gender ProblemNow. The Atlantic, March 16, 2017. https://www. theatlantic.com/business/archive/2017/03/melinda-gates-tech/ 519762/ (accessed March 26, 2018). (4) U.S. National Science Foundation. Women, Minorities, and Persons with Disabilities in Science and Engineering. https://www.nsf. gov/statistics/2017/nsf17310/data.cfm (accessed March 25, 2018). (5) Catalyst: Workplaces that Work for Women. Women in Science, Technology, Engineering, and Mathematics (STEM). http://www. catalyst.org/knowledge/women-science-technology-engineering-andmathematics-stem (accessed March 25, 2018). (6) LeanIn.org and McKinsey & Company. Women in the Workplace 2017. https://womenintheworkplace.com/ (accessed March 25, 2018). (7) The current Gordon Research Conference on Computer-Aided Drug Design was originally titled Quantitative Structure−Activity Relationships in Biology and then Quantitative Structure−Activity Relationships. The name was changed to Computer-Aided Drug Design in 2001. (8) Huryn, D. M.; Bolognesi, M. L.; Young, W. B. Medicinal Chemistry: Where Are All The Women? ACS Med. Chem. Lett. 2017, 8, 900−902. (9) A recent study demonstrated that biology, chemistry, and physics professors continue to rate male students as more competent and hirable than female students applying for the same undergraduate science laboratory manager position. Additionally, the mean starting salary offered to female applicants is significantly lower than that offered to male applicants. See: Williams, W. M.; Ceci, S. J. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 5360−5365. (10) Banajii, M. R.; Greenwald, A. G. Blindspot: Hidden Biases of Good People; Delacorte Press: New York, 2013. (11) Huggett, B. Biotech’s Pale Shadow. Nat. Biotechnol. 2018, 36, 20− 30. (12) MARC: Men Advocating for Real Change. http://onthemarc. org/home (accessed March 26, 2018). (13) Jarvis, L. Why Can’t the Drug Industry Solve Its Gender Diversity Problem? Chem. Eng. News 2018, 96 (10), 26−33. https://cen.acs.org/ articles/96/i10/why-cant-the-drug-industry-solve-its-gender-diversityproblem.html (accessed March 26, 2018). (14) Bonham, K. S.; Stefan, M. I. Women are underrepresented in computational biology: An analysis of the scholarly literature in biology, computer science, and computational biology. PLoS Comput. Biol. 2017, 13, e1005134. (15) This was a New England Council slogan in the 1960s that was coopted by Ted Sorenson for a John F. Kennedy speech.
E
DOI: 10.1021/acs.jcim.8b00170 J. Chem. Inf. Model. XXXX, XXX, XXX−XXX
