Mentoring philosophy in the Tanigawa Lab

Wed, 29 Apr 2026 00:00:00 +0000

I work with students, postdoctoral scholars, and visiting researchers across career stages. Although each person’s background and long-term goals are different, I strive to build a research environment that is supportive, rigorous, collaborative, and growth-oriented.

I also grow by teaching and mentoring others. As such, please consider this a growing document of my teaching and mentoring philosophy.

Mutual fit matters

A mutual fit between the advisor and the trainee is critical to ensuring that your time in the lab is productive and rewarding. I encourage prospective lab members to think carefully about their long-term goals, learn about our research, and assess its alignment with their interests and strengths.

When we talk, I want to learn what you have done. I also want to understand what kinds of problems excite you, what skills you want to build, and how I can support your development.

My goals for scientific training and mentoring

I believe research and associated analytical skills will enrich your life. When you are lucky in research, you may discover something that nobody else has seen before. At the very least, you will cultivate an ability to engage with subject matter of your choice using multiple perspectives. However, learning those skills takes effort, even with the help of advisors, colleagues, or AI. It often takes the form of an iterative process.

It will be rewarding for some, but I also understand that not everyone will enjoy the iterative process. In classroom settings, students often engage with defined problems, often with known solutions. In contrast, in research settings, it is often unclear what exactly the problem is.

As an educator and mentor, my goal is to train next-generation leaders in biomedical data science and related fields. I have somewhat different objectives at different career stages. For PhD students and postdocs, I expect them to lead independent research programs. This means that PhD students, especially senior PhD students, and postdocs should be able to clearly articulate the types of problems they are working on, why those topics are exciting and worthy of attention, why they are well positioned to lead such a study, how their findings are supported by the data, and why the work matters to the broader scientific community and/or the general public. That requires synthesizing many skills, and I will work with each individual to identify milestones in their learning trajectory and help them develop the skills they need to succeed in their careers.

For master’s and undergraduate students, having exposure to the research process is a primary goal of the mentorship. Some students may enjoy doing research and aspire to pursue a PhD, and I would be excited in that case. Others may not see PhD training as an attractive path. That is totally okay. I am grateful for the mentors who gave me research opportunities as a student. Some skills are better taught in research settings, and within our advising bandwidth, I would love to pay that forward.

Of course, grouping people by career stage is not a perfect way to address individual needs. When trainees join my group, I will use individual 1:1 meetings to assess their goals and alignment of interests. The description above is meant to give some idea of how I approach mentoring in research settings.

Research environment and professional development

The lab provides a computational environment and a strong network for collaboration with experts across human genetics, computational biology, and biomedical data science. I have extensive experience analyzing biobank-scale data, integrating omics profiles, and leveraging those findings for therapeutic target discovery.

We strive to create a highly collaborative, inclusive, and supportive environment that fosters growth and success. As one of the early members of a new lab, you may also receive especially direct mentorship and the chance to help shape lab culture.

In terms of infrastructure, the lab has a lab wiki for documentation and lab notebooks, Slack, and high-performance computing. I am happy to invest in whatever resources help you grow as a researcher, of course, within reasonable resource constraints.

Mentorship is a two-way street

Good mentoring relationships rest on commitments from both sides. Below, I describe what I aim to provide as a mentor and what I expect from trainees in return.

How I support trainees

For your training, I strive to provide the mentorship, support, and resources needed for professional success. That includes regular scientific discussion, direct feedback on project development, and opportunities to build independence over time.

In practice, trainees can use weekly meetings with me to address questions, receive feedback, and discuss next steps. I also value peer support through lab meetings, Slack, and social gatherings. As people progress, I try to provide opportunities for presentations and professional networking, including conferences and interactions with collaborators.

Expectations and scientific growth

At the same time, I expect trainees to take initiative and ownership of their education and research. Methodologies for scientific research and scientific communication are learnable skills, and I want to help people develop them deliberately. I will use a combination of weekly meetings, Slack messages, and end-of-quarter reviews to communicate my expectations and stay aligned on your scientific development.

Can you picture yourself, a few years from now, applying statistical models and computational tools to address a problem that hooks you? Would you be excited to design your training plan to reach that stage, with help from advisors and peers?

I tend to work best with individuals who have strong intrinsic motivation. Depending on the lab’s advising capacity and the fit between your background and our active projects, I may recommend that you first strengthen specific technical foundations through coursework or independent study before starting a research project.

What is it like to work with Yosuke?

Initial interactions

When you engage in a conversation with me, I will often ask what your specific interests are and what motivates you to work with me. There are many resources out there, and a broad statement like “I’m interested in AI for Health” gives me the impression that you are still exploring the field as a whole. At that stage of learning, taking classes or watching online lectures would be more appropriate so that you can make an informed decision about what to investigate in your research project. Instead, if you can communicate specifics, such as a paper that excited you, a question you keep coming back to, or a method you want to learn, that is a promising signal. Ultimately, I’d like to know what keeps you up at night, or what excites you every morning. If that aligns with what we care about as a group, I think there is a beautiful synergy, or at least good potential for it.

I have a set of questions I prefer to ask applicants, and together we will assess the alignment of interests. We will also identify potential gaps in prior training and see whether I can provide resources to help close them.

Getting started in the lab

Once you join the group, we will most likely have weekly meetings. At the beginning, I often ask you to read introductory, review, and state-of-the-art research papers. I may not give you “copy-and-paste-ready” scripts at the beginning because I believe it is critically important to expose you to the field’s literature so you develop the learning skills that enable you to catch up on the most exciting topics. We will use weekly (or more frequent) meetings to resolve your confusion. I am happy to answer your questions and provide additional resources. When there is a good foundation, I also ask you to read one or two papers in detail. Attention to detail is an important skill in research. Going through the details, evaluating your level of understanding, and articulating areas of confusion are all very important.

Research in the lab

When you are ready, you will start developing models and analyzing data. The emphasis here is that you need to understand the biomedical and statistical meaning of each computational operation. Simply because a colleague or an AI gave you some snippets does not mean they are the right strategy to pursue. We will discuss your results and progress via weekly meetings, Slack, and casual chats. Those updates are excellent opportunities to cultivate communication skills.

Scientific communication

When you have a substantial body of results, you will be asked to write them up as a manuscript. It can take time to produce a first submittable draft as a first author. That was certainly true for me. I will point you to relevant writing resources and help you acquire those skills. It may sound daunting to see repeated loops of my feedback. Through many rounds of writing, submission, and revision, I have built up a sense of what the final deliverable should look like, and I lean on that calibration during this stage. From experience, polishing the manuscript before submission is much easier than fixing major problems after the fact. When the time comes, we will submit it to a journal, revise it, and hope it is accepted for publication.

When you have a manuscript submitted, you may have the opportunity to present your work at conferences. Those decisions are made on a case-by-case basis, depending on the type of work, the kind of conference, the expected audience, the distance from the lab, the duration, scheduling constraints, and so on.

Publishing your paper

I hope you will be able to publish your work. The sense of accomplishment from the entire process can be tremendous. By that point, you will have identified a gap in humanity’s knowledge, devised a thoughtful strategy to address it, filled that gap, and satisfied your peers to a reasonable standard. You will have added a new piece of information to human knowledge, which may be cited and used by many others.

Who is likely to enjoy working with our group?

Research takes time and effort. There are moments that require dedicated effort, such as conference deadlines or degree progress milestones like the oral defense or thesis submission. In our field, long hours do not necessarily lead to better outcomes. Your computational analyses may give you sensible results on the first try, or you may need to iterate 10 times. We each need to find our own balance. Having a north star will support you in challenging times. If you lose sight of why you are engaged in research, it will be difficult to overcome challenges.

I’d like to work with someone with genuine intellectual curiosity, internal drive, and a growth mindset. It took me years to become a researcher with reasonably strong research skills, thanks to my mentors’ patience. Doing good research, practicing good science, and mentoring well are all ongoing processes.

📣 Recruiting

We are recruiting at all levels. Please check the Join Us page for more information.

Success of past trainees

Over time, I have mentored around 20 students spanning high school through graduate training. Former informal mentees have presented at major conferences, received awards, and moved on to competitive positions. You may see some examples of trainee presentations at the . I often introduce current trainees to former mentees, so you can see role models walking a few years ahead of you.

Learn more

If this mentoring style resonates with you, you may also want to look at our , , and pages.

I also recommend the following resources for thinking about graduate training and research:

by Phillips and Johnson, a general guide to graduate training and research habits.
by Uri Alon, an essay on identifying research problems worth pursuing.

Research Training | Tanigawa Lab