I didn’t have the time or resources

to enroll in a degree program. So I

would need to be creative. Thank-

fully, my postdoc adviser agreed

I could spend some time to learn

bioinformatics on my own, even

though he already paid someone to

do that task. I persuaded him that

someone with lab experience could

bring new insights to the data

analysis process. But I still had ex-

periments to run and animals to

care for, so I had to be strategic.

I looked over my calendar. It

didn’t have a lot of wiggle room;

finding time to learn something

new was going to be a challenge.

I re-evaluated every entry and

thought hard about how much

time I was spending on each task.

I was attending several semi-

nars and journal clubs each week,

which were not bringing much

value to my work. I was also spending way too much time

preparing for lab meeting presentations, trying to perfect

my slides. And I was doing tasks in the lab that I could

reasonably delegate to others.

Our lab had hired a new research assistant who was

working on my project. She had limited experience, but

after I invested time to train her, she was able to take over

many lab tasks, which freed up a lot of my time.

The next step was to sit down and learn. I began by

working with a data set I’d generated in the lab—one that

I would have normally passed off to the bioinformatician

to analyze. I started by learning just enough to complete

small tasks, writing a few lines of code before stopping

to look up what I needed to do for my next step. I relied

on online resources, and I asked our bioinformatician for

help when I got stuck. It was a steep learning curve, but

by the end, I’d completed an analysis that we included in

our publication. This gave me the confidence to take my

learning up a notch.

Over time, as I analyzed more data

sets and learned new skills, the im-

poster syndrome voice in my head

began to diminish. Coding quickly

became an obsession; I impulsively

executed command after command

just to see what they do.

After 3 years, I applied for a post-

doc position in a lab that didn’t have

a bioinformatics expert on hand al-

ready. I was recruited because of my

skills with large genomic data sets,

and I saw the position as an oppor-

tunity to further hone those skills. It

helped me do that. But more impor-

tantly, it taught me what I wanted in

a career.

I was still tasked with some bench

work during my second postdoc,

and I often found myself frustrated

that I was not able to spend 100% of

my time on computational work. I

started to find lab tasks monotonous

and I grew impatient during rate-limiting steps—for instance,

when I was waiting for cells to grow. In contrast, when I was

working on a coding problem or a data visualization chal-

lenge, my creative juices flowed. I also produced results at a

faster rate, which made me feel productive. That helped me

see that, going forward, I only wanted bioinformatics jobs.

I started to apply for faculty jobs and in less than 2 months

I secured a position to lead a data science group. I’ve been in

that position now for almost 3 years, and I’m happy I made

the switch away from bench science. I now focus on all things

data science, whether in teaching or research. And I feel free

and in control of my time.

Curiosity initially enticed me to try my hand at bio-

informatics. Over time, it grew into a full-fledged career. It

wasn’t easy to learn a new skill set while I had experiments

to run and papers to write. But taking the time to do that

opened doors for me and led me to a career I truly enjoy. j

Alvina G. Lai is an associate professor at University College London.

“Coding quickly became an

obsession. ... It taught me what

I wanted in a career.”

Following my curiosity

I

’m a lab biologist by training. I’m good at running experiments, taking care of animals, and cloning

difficult genes. But as a postdoc, I started to crave a new challenge. I began to look over the shoul-

der of the bioinformatician whose job it was to wrangle and analyze gigabytes of genomic data that

our group generated. It was complicated and challenging work that required a unique skill set,

which was why we had a person dedicated entirely to the task. But after learning about his work

and seeing his code and visualizations, I grew interested in learning how to do it myself. With

faculty positions hard to come by in my field, I figured, “What better time to upskill than now?”

By Alvina G. Lai

ILLUSTRATION: ROBERT NEUBECKER

690 11 FEBRUARY 2022 • VOL 375 ISSUE 6581 science.org SCIENCE

WORKING LIFE