Skip to main content
- Open communication. Science is complex, and especially in interdisciplinary science, it often ends up being the case that one person is talking about a topic on which some others in the room have no idea. Not in this lab. We are adamant that each person speak out when they don’t follow, and ask when they don’t understand. We call out pretentious ignorance, not asking questions. Another corollary of the ‘no bull’ rule: every word that comes from a team member’s mouth must be supported by a study in the literature, an analysis they have run, or another piece of data. Pure speculation is called out and rooted out.
- Do the best. We first ask ‘how do we best solve this problem?’ and then do it. We are agnostic to how easy or difficult that road is. Forget the ‘low-hanging fruit’; there exists no such thing in cancer-drug discovery.
- Work hard. We are a handful of young scientists trying to change the world’s drug discovery; that won’t happen easily, and we are fully aware of that.
- Machine learning = statistical inference. We follow machine learning with theoretical guarantees and ignore/frown upon the use of fuzzy logic, genetic algorithms, and other methods that lack such guarantees. We are firmly committed to a view of machine learning as being essentially statistical inference with a fancy name.
- Training is key. Cancer biology cannot really be studied while ignoring immunology. Machine learning cannot be studied without an in-depth understanding of statistics. Most new graduate students/new postdocs don’t come in with every skill that is necessary for them to excel in their fields (see rule #1). That is okay. We first see what skills the new member needs to do the best they can do (see rule #2). Then give them the time, resources, and environment for that training. Then we expect them to pull their weight and actually learn those skill sets (see rule #3). If it takes a year to give that training, so be it. Our philosophy is that investing in training will yield higher returns than its cost.
- Rock hard. We often listen to hard rock when working, but we can put on headphones if necessary.