# Monthly Archive: October 2017

## Remembering Mimi

Seventeen years ago a driver at Bill’s company was headed south on Route 2 towards Glastonbury. He swerved to miss a paper bag in the road and because he didn’t want anyone else to damage a tire running over a bag of beer bottles, he pulled over and retrieved the bag. Instead of beer bottles, he found two tiny black kittens inside. He was late for a delivery, but he returned to the office and left them there. What follows is a very long story, but at the end of the day Bill took them to our veterinarian. They were so tiny he didn’t know what they could eat. The vet gave Bill some kitten formula, and he brought them home. They were so close to starvation that their bodies shook as they lapped up the formula. They were so small that I could easily hold both of them in the palm of one hand. It was several weeks before they were big enough to climb the stairs to the second floor of our house.

By Sunday night we had named them. Mimi because she was so sweet and kind, as in Puccini’s La Botheme. Maxwell (“Max”) because he is a lovable and ungainly like Maxwell Smart.

Yesterday we lost Mimi. Her health had been declining for more than a year, but in the past week and a half the decline became precipitous. There was a sadness in her eyes, and she was happy only when one of us held her. She passed away peacefully a little after 4:00pm. Max is still healthy, but the three of us, including Max, are heartbrokn. Our house feels empty, but Mimi’s suffering is over. She lives on in our memories, and we will hold her close to our hearts forever.

## Championing the Success of Women in Science, Technology, Engineering, Maths, and Medicine

In honor of Ada Lovelace Day, the second Tuesday in October, Digital Science released a report entitled Championing the Success of Women in Science, Technology, Engineering, Maths, and Medicine. I encourage you to read it, and not only because Lauren Kane (COO of BioOne1) is a co-author of one of the chapters. Here how the report is described on its Figshare page.

This report explores the role of women in STEM and the challenges they face, looking at areas of gender inequality, exploring potential causes of this inequality and offering solutions. Women’s reluctance to step into leading roles, their tendency to suffer from “imposter syndrome” and their career breaks as a result of motherhood, are just some of the contributory factors holding them back, as well as the outdated, sexist attitudes they sometimes have to face in the workplace.

## Plants, People, and the Mother City

Tanisha Williams, Fulbright 2015-2016, South Africa, at Boulders Beach visiting the penguins.

Some of you know that Carl Schlichting and I co-advise Tanisha Williams. If you know that, you almost certainly know that Tanisha spent the 2015-2016 academic year as a Fulbright Fellow in South Africa. She was based at the Cape Peninsula University of Technology, and she used her time not only to collect seeds of Pelargonium and establish experimental gardens at Kirstenbosch Botanical Garden and Rhodes University but also to work with two non-profit environmental organizations. She posted an article about her experience on the blog of the Fulbright Student Program. Here’s an excerpt to whet your appetite:

Among the many experiences I had, I must say the residents from the Khayelitsha township have taken a special place in my heart. This is where I taught girls and young women math, science, computer tutoring, life skills, and female empowerment through a community center program. It was such an impactful experience, as these girls are growing up in a community with high rates of unemployment, violence, and other socioeconomic issues. It was empowering for me to see the curiosity and determination these girls had for learning and changing their community. They thought I was there to teach them from my own experiences being raised in a comparable situation and now working on my doctorate as a scientist, but I know I was the one that gained the most from our time together. I learned what it truly means to have hope and persevere. These lessons, along with the ecological and evolutionary insights from my academic research, will be ones that I always remember.

## Exploring mixed models in Stan

I am about to begin developing a moderately complex mixed model in Stan to analyze realtinoships among anatomical/morphological traits (e.g., leaf thickness, LMA, wood density), physiological performance (e.g., Amax, stem hydraulic conductance), and indices of fitness (e.g., height, growth rate, number of seedheads). One complication is that the observations are from several different species of Protea at several different sites.1 We’re going to treat sites as nested within species.

Before I start building the whole model, I wanted to make sure that I can do a simple mixed linear regression with a random site effect nested within a random species effect. In stan_lmer() notation that becomes:

stan_lmer(Amax ~ LMA + (1|Species/Site))


I ran a version of my code with several covariates in addition to LMA using hand-coded stan and compared the results to those from stan_lmer(). Estimates for the overall intercept and the regression coefficients associated with each covariate were very similar. The estimates of both standard deviations and individual random effects at the species and site within species level were rather different – especially at the species level. This was troubling, so I set up a simple simulation to see if I could figure out what was going on. The R code, Stan code, and simulation results are available in Github: https://kholsinger.github.io/mixed-models/.

The model used for simulation is very simple:

$$y_k \sim \mbox{N}(\mu_k, \sigma) \\ \mu_k = \beta_0(species|site) + \beta_1x \\ \beta_0(species|site) \sim \mbox{N}(\beta_0(species), \sigma_{species|site}) \\ \beta_0(species) \sim \mbox{N}(\beta_0, \sigma_{species})$$

Happily, the Stan code I wrote does well in recovering the simulation parameters.2 Surprisingly, it does better on recovering the random effect parameters than stan_lmer(). I haven’t completely sorted things out yet, but the difference is likely to be a result of different prior specifications for the random effects. My simulation code3 uses independent Cauchy(0,5) priors for the standard deviation of all variance parameters. stan_lmer() uses a covariance structure for all parameters that vary by group.4 If the difference in prior specifications is really responsible, it means that the differences between my approach and the approach used in stan_lmer() will vanish as the number of groups grows.

Since we’re only interested in the analog of $$\beta_1$$ for the analyses we’ll be doing, the difference in random effect estimates doesn’t bother me, especially since my approach seems to recover them better given the random effect structure we’re working with. This is, however, a good reminder that if you’re working with mixed models and you’re interested in the group-level parameters, you’re going to need a large number of groups, not just a large number of individuals, to get reliable estimates of the group parameters.

## Announcing the BioOne Career Center

BioOne is a collaboration between libraries and non-profit scholarly publishers in organismal and environmental life sciences. It was founded in 1999 to help publishers obtain the revenue they need to support their publishing program while ensuring affordable access to scholarly journals for libraries and their patrons. I am proud to have served as Chair of the BioOne Board of Directors since 2000.

BioOne’s primary service is to provide BioOne Complete, a database of 207 journals including many open access titles. As the title of this post suggests, BioOne is now offering a new service, the BioOne Career Center. Anyone looking for an opportunity can create a free account, set up a job aloer, and post their CV. Employers can post jobs on the site for free until the end of October (you’ll find the necessary code in the announcement), and posting for internships, volunteer opportunities, and conferences will always be free. We hope that the BioOne Career Center will become a valuable resource.