Assistant to Robot, Promoted to Robot

I was telling my grad students this story last week, and they liked mocking me so much as "assistant to a robot" that I thought I should post the story on here so more people can mock me. My first job in a research lab was the summer before starting my undergraduate career at the University of Michigan as a physics major--1992. I was really lucky get a summer job in one of Francis Collins' labs at UM. Yes, I am name dropping. The name I just dropped was that of Francis Collins, who was leader of the NHGRI from 1994-ish to 2008. Prior to that he was at the University of Michigan, with primary roles of hiring me as a work-study student and also leading teams that found the genes for cystic fibrosis, Huntington's disease, neurofibromatosis, and other diseases.

I have been lucky so often in my life, and in particular in my career "planning." I'll tell you some other day how lucky I got in obtaining my current job at UNM. This is how I obtained my first job in a research lab: I was friends with Dr. Collins' daughter, and I liked science. I knew he had a research lab because he had visited our classroom in Junior High to tell us about cystic fibrosis and genetics. So, I asked my friend if I could work in her Dad's lab. A few days later, she told me, "he says yes," or something along those lines. I was 17 years old at the time. But when writing this story, it seems like I was younger, as I recognize this strategy as the same one I used for obtaining a rollerskating "skate" with a girl in the 6th grade. I think the song was "Manic Monday."

I actually worked in a lab led by Chandra Sekharappa, who I think now has this lab. He was such a great guy and I am eternally grateful to him, Dr. Collins, and the other people in that lab who welcomed the unusual physics undergraduate to their lab. As I am writing this blog entry, a flood of memories are coming back to me. I learned so many things from working in this lab, and now, 17 years later, they are still coming back to me and helping me in my research (which coincidentally, or probably not coincidentally is tending towards genomics applications). In this lab is where I learned to pipette. I learned what PCR was. I unfolded paper towels for Northern blots. I "stuffed tips" (FYI: I could use each hand independently on two different boxes). I helped with "rows and columns." I washed dishes. Wow, did I wash dishes. I became obsessed with: -80 freezers; dry ice; vacuum-bake ovens; centrifuges; liquinox; reverse-osmosis water; latex gloves; latex gloves filled with water and frozen in the -80C freezer; and latex gloves in the vacuum-bake oven.

I cannot even come close to expressing how important this experience was to my career. Being immersed in this environment was so valuable -- whether I knew it at the time or not. The lab was focused on cloning the gene for early-onset familial breast cancer. (I believe another lab ultimately beat them by identifying BRCA1, but I'm not sure.) There was such a palpable excitement about the race to find this gene and I loved watching it. I distinctly remember that Dr. Collins welcomed me into group meetings, where the postdocs or grad students (I'm not sure what they were) would pass around these developed images of gels with the faintest of bands that proved something about their PCR reaction. I distinctly remember that they'd let the ignorant physics undergraduate stare at the film and then tolerate it when I said, "I think you're crazy, there's no band there." Somehow they kept inviting me, and they kept trying to explain to me "gene jumping" or "chimerisms" or "FISH" or other topics. The collaborative atmosphere in Chandra and Francis's lab is something I'm striving to replicate in our lab at UNM.

OK, now onto the good part. Of course, being the undergraduate in the lab made me a target for the grunt work. More than that, I wasn't even a biologist! So, it happened that the lab (or someone on the floor) had gotten their hands on a robotic system that could essentially print microarrays on filter paper. Or perhaps the predecessor to microrrays. The "robot" could print media from sixteen 96-welled plates onto a single filter paper. Then, these 4x4 arrays could be used for some kind of hybridization assays. This was a big deal, and the robot cost something in the 100's of thousands of dollars. Basically, you'd put a stack of 16 microtiter plates in the holder next to the robot. You'd set up the filter paper, and then the robot would proceed to: grab a plate; take lid off plate; put plate down; stick pins in plate; stick pins on filter paper; clean pins; put lid on plate; put plate away; repeat with new plate.

The problem was, the robot was controlled by some kind of SGI machine that nobody knew how to program. It cost a whole bunch of money to have the tech rep come out and program the thing. Everything about the robotic system worked well. Except, after taking the lid off the plate, it accelerated too fast, and media would splash from one well to another. This was terrible. I know what you're thinking: ask the physics undergrad to reprogram the robot! This is what I was thinking too when the grad students (or postdocs) explained to me the problem. I'm pretty sure I could have figured this out, no matter how obscure and proprietary the programming language. But, this was not my fortune. Instead, what they had figured out was that my $5.50 / hour salary was a perfect solution. I could perform the first part of the robotic sequence (grab plate; take off lid) and then at the appropriate time, hand the plate to the robot. So, this is when I took on my esteemed position as "assistant to the robot." I don't know how many days this lasted ... probably not too many, I think maybe for a few hundred plates or so. I do remember how utterly boring it was. I actually tried to read a book in 20 second increments while I tag-teamed with the robot.

Perhaps during my time as Assistant to the Robot, I impressed people enough to get my first promotion in the lab: to actual robot. (I previously mentioned my prowess at stuffing tips and unfolding paper towels, which probably factored into this promotion.) This job took most of a summer (1993 maybe?) and actually I'm pretty proud of it. My task was to copy the Washington University YAC (yeast artificial chromosome) library. I think it was about 200 96-welled plates and it took me most of a summer to make two copies. I became the most efficient plate-pourer of all time (in my own mind), and discovered that you can actually pour them so thin that even yeast can't grow. I wonder if these YAC libraries are still around nowadays?

Well, that's the anti-climactic ending to my story. I don't have a coherent point, and I know this goes against all of the how-to-be-a-good-blogger advice. My points are: (1) I collaborated with a robot in the past because it was cheaper than fixing the robot and (2) I had an awesome undergraduate research experience that has profoundly impacted my career. In regards to (2), there are so many lessons I can learn to help me in my current position as a research mentor. The main thing I have been thinking is that undergraduate research can be valuable for the lab, and incredibly valuable for the undergraduate. I feel like we're not even coming close to achieving what we could at UNM in regards to undergraduate research, and I would like to change this over the next couple years. I routinely meet Junior-level physics majors who are interested in research, but haven't yet been in a lab. We are next door to Sandia National Labs, and only 2 hours away from Los Alamos National Lab...both of which have amazing resources and opportunities for undergraduate scientists. And of course, we have plenty of our own labs at UNM. One of my goals over the next few years is to help our students find research jobs earlier in their careers...perhaps even before they start at UNM. Whether their jobs can be as prestigious as my own assistant to robot jobs, I don't know, but I can definitely strive for that!

SJK Note 4/2/09: I found a picture in my garage of the completed robot project. That's me admiring my 400? or so microtiter plates, all nicely stacked and labeled.

A science outreach idea, what do you think?

I live on a cul-de-sac where we are lucky enough to know and enjoy hanging out with many of our neighbors. Many have kids who play with our kids. It's like when I was a kid, and I thought those days had passed, but they haven't. (As an aside, I love hyphen-ating words, but it bothers me that cul-de-sac is hyphen-ated.)

Many of my neighbors really enjoy hearing about the science we're doing in the lab, and I really enjoy talking about it with them. This actually led to a very fun event we did during Winter Break where I brought in some neighbors to our Junior Physics lab course so they could get hands-on experience with some very cool physics. You can check out our OpenWetWare page (unfinished) for the event (sorry the facebook page seems to be private). A brief summary is that I only had to invest a few hours of time, and I think the attendees really enjoyed it. I know I did.

Recently I had an idea for science outreach that I'd like your opinion on. The idea is that I (or a student) will explain our research to one of our biggest neighbor fans. Then, we'll record an interview with him describing our science, what we do, it's importance, etc., from his point of view. Or we could do it with a couple neighbors talking. But the main point is that the non-scientists will be explaining the science to the (mostly) non-scientist audience on youtube.

There are a few reasons I think this may be useful and fun. First, I always find it informative and fun to hear people "re-describe" our research to someone else after I've described it (unless it's printed in a magazine). Second, I have an inkling that it would be effective for communicating to non-scientists. Third, the people I have in mind for this project are very good at picking out the essence of what I'm telling them, and distilling it down to the exciting parts in layman's terms.

So, do you think this is a good idea? Maybe it's been tried before many times, and if so, please send me the links. If we do give this a whirl, there is one thing I'd like to figure out how to do:

• Record video with two cameras (for example, one on me, one on him)
• Splice and edit the video to make a good video for posting to youtube

I'd really appreciate advice on software and hardware to use for those purposes. Thanks!

Personal open science challenges

There was recently a very interesting thread regarding open notebook science in the Science 2.0 friendfeed room. This was in response to Michael Nielson announcing that Tobias Osborne had begun doing open notebook quantum information theory. I think this is fantastic, and my kudos go to Tobias (whom I don't know). The friendfeed debate had to do with whether Tobias's work can be called open notebook science, which has a specific definition.

The debate got me thinking again about something that's been bothering me recently. I've been having a hard time getting my thoughts straight, and that's still true. I'll quote myself and then try to clarify:

A really good motto for a scientist who wants to be open could be this: "Be as open as I personally want to be." This is very different than "be as open as possible." What I am specifically thinking is that young scientists (i.e., not yet beaten-down) seem to usually have very natural tendencies towards open science. But the overall level of natural talent for openness may vary enough that "open notebook science" may just not be the best method of openness for some people. But everyone can strive to "be as open as they want to be", and resist pressure to be closed coming from outside (fear of scooping; lack of technical means; resistance from colleagues). In contrast to these external pressures, I think it may be legitimate for someone to want to be open, but also maintain some privacy so they can get a personal reward of doing something all by themselves, for example. Perhaps posting all of their electronic notes 6 months or a year down the line.

"Be as open as I want to be." I don't know if that has value for anyone else, but it a very powerful mission statement for me right now. It's powerful, because I really believe in it, but I am not achieving it. I'll talk about that later in the post. But, first I want to talk about it in a more positive light.

What kind of openness should be required?

I am starting to decide that I'm not going to try to force my lab members to do specific kinds of open science. I am thinking instead that my goal will be to remove as many barriers as possible so that my lab members can achieve the level of openness they desire. I believe that adults have unchangeable natural talents, and I think that scientists will be cutout for different kinds of openness. For example, Anthony in our lab has recently started doing open notebook science, true to its definition. I am really excited about this. He is a natural for ONS. I don't think that he has any problem writing anything in public. In fact, I think his notebook being open is a motivator for him to make it even better than he would a private notebook. This is the way he's wired, and it's not surprising if you know him. In contrast, I think some people would find that their creativity and drive are seriously hampered by doing ONS. For example, me as a graduate student. I don't know whether doing ONS would have worked or not. I actually kept what I think is a very good electronic lab notebook. But it was private, and I don't know whether I would have taken as many notes (and dropped as many F-bombs) if I knew it was public. I also don't know if I would have reacted well to someone posting a suggestion to me when I was immersed in trying to figure out something by myself. I do know that I would have been fine posting my notebook in public with some time delay. In fact, if anyone posts a comment to this blog asking me to post my grad school notebook in public, I'll go ahead and do that...f-bombs and all.

So, while I don't think I'll require ONS for all lab members, I may have other requirements, such as delayed notebook publishing. What I am worried about is hampering creativity and productivity of young scientists by striving for inappropriately selected open science goals. I do want my students (and postdocs in the future) to strive for open science, but I want them to do it in the way that best leverages their talents.

I am failing at my own principles

"Be as open as I want to be." I and our lab have made some great strides in the past few months towards this principle. For me, I think the transformation was fueled by a strong belief in the power and even morality of open science. But it did take a heavy dose of "what the fuck" to spark the flurry of steps I took this past winter break. (I think that may be my first f-bomb while blogging; I feel alive.) I'm happy and excited about what we're doing. But I'm also not achieving openness as much as I'd like. And I'm confused. Two themes dominate my struggles with openness:

• The students in my labs and their scientific careers
• My collaborators, their careers, and my gratitude for their assistance and mentoring
  

I'm not trying to sound altruistic here. One of my talents is that I get genuine happiness out of feeling like I've helped other people succeed. You can see that both of those items above feed that desire in me. I do think those two items are what is confusing me. In contrast, the issue of being scooped, in itself does not impact my thinking. I do worry about being scooped, but I have already concluded that being open does not increase the chances of being scooped. I believe being open decreases the chances of being accidentally scooped substantially. Furthermore, I even believe that being completely open would reduce the chances of being purposefully scooped. This is because the published track record would make it easier to shame the person who did the scooping.

Being scooped would be emotionally devastating. This is true. And it would have an impact on my lab and my students. This is what my students and I have been discussing the past couple years, and I think we've developed a collective (perhaps unspoken) understanding that we'll be OK even if that does happen. I think I can protect and rescue my students from that scenario. The collaborator issue is so much more complicated.

The collaborator issue is what is bothering me quite a bit now, and I really don't have any answer. Most of the scientists I know personally are "traditional." The ones I am trying to collaborate with are outstanding and highly respected by everyone, including me. The ones I am thinking about right now have put in a huge amount of effort helping me throughout various stages of my career. These traditional scientists, of course, are not Scientists 2.0, but they are fantastic scientists. I suspect, and in some cases directly know, that they would not approve of my science openness. So, I don't know how to deal with this external pressure towards closed science. The "what the fuck" strategy seems so disrespectful to people who've put energy into my career. But the "try to convince" them strategy is futile. "Showing them the way" will work...but at the risk of looking like "what the fuck" along the way and angering them. If we do get scooped, my students and I will be OK. But our mentors may never forgive us?

OK, I'm going to stop now...those are the challenges that are really bothering me this weekend.