How to Choose a Bioinformatics PhD Supervisor

Choosing a bioinformatics PhD supervisor means judging a person, not a topic. Read the papers where they are senior author, check how many scholars they already supervise, ask current students specific questions, and read past students’ thesis acknowledgements. Indian regulation requires institutions to publish both lists, so most of this is checkable before you email.
By this point you probably have a shortlist of groups and a draft email you have rewritten four times. Our guide to applying for a bioinformatics PhD in India and abroad covers eligibility, entrance tests and the mechanics of the application, and the research proposal guide covers the document. Neither covers the decision that outlasts both of them, which is the choice of person. This page starts at your shortlist and ends when you press send.
What you will find on this page
- What a supervisor actually controls during a PhD, and why that makes the person a bigger decision than the topic
- How to read a publication record in a few minutes, with a query you can run today and re-run for any name
- What India’s UGC regulations say about who may supervise you, how many scholars they may already have, and which of that the institution must publish
- The four public records you can pull before making contact, compared in one table, with what each one cannot tell you
- What to ask current and past students, and how to read an answer that is carefully polite
- Red flags, interview questions that go both ways, and an FAQ
What does a PhD supervisor actually control?
More than the subject of your thesis. In the Indian system the topic is not even fixed when you choose the person: under clause 10(1) of the UGC regulations, a Research Advisory Committee reviews your proposal and finalises the topic after admission. So the thing you are deciding now is not what you will study. It is who will decide, with you, how the next four to six years run.
Three of those controls matter more than students expect. The first is money, both the project kind that buys compute and the personal kind that pays your rent. The second is time: how often you meet, and how long a draft sits before it comes back with comments. The third is professional development, meaning whether mentoring juniors, teaching, an industry placement or outreach counts as part of your training or as a distraction from it. “Ten simple rules for choosing a PhD supervisor”, published in PLOS Computational Biology in 2021, puts expectation alignment at the centre of the decision and tells students to raise project funding and stipend early rather than treating finances as impolite.
There is also a control that is not negotiable at all. If a faculty member is within three years of superannuation, Indian regulation stops them taking a new scholar, however warm their reply to your email. That is worth knowing before you invest a month in a conversation.
How do you check a supervisor’s publication record before you email?
Start with their papers, not their profile page. The PLOS paper is blunt about group websites being frequently out of date, and recommends the publication profile and students’ theses instead. It also gives two reading rules that change what you see. Pay particular attention to papers where the prospective supervisor is senior author, which in the life sciences usually means last author, because those map where the group is going rather than where a postdoc took it. And remember that publishing is slow, so this year’s papers may describe work finished two years ago.
Europe PMC lets you count a record in one command, with no subscription. The following returns the number of indexed records for an author string:
curl -s -H "Accept: application/json" \
"https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=AUTH%3A%22McCain%20JSP%22&format=json&pageSize=1"Run on 17 November 2026, that returned a hit count of 16 for one of the PLOS paper’s own authors. Adding a year filter shows whether a group is currently active rather than historically known:
... &query=AUTH%3A%22McCain%20JSP%22%20AND%20PUB_YEAR%3A%5B2024%20TO%202026%5DThat returned seven. Two counts, one query, and you now know something a profile page will not tell you. Two warnings come with it. A surname plus initials is a blunt instrument: it merges different researchers who share both, and misses papers published under a different initial or a changed name, so treat the number as a starting point and open the titles. And a zero means nothing on its own, because a malformed query returns zero exactly as an unproductive author does. Always run a name you know works first, as a control.
For a cleaner identity, use ORCID, which gives each researcher a persistent identifier that follows them between institutions. The registry is large enough that most active researchers are in it: a public search for every record, https://pub.orcid.org/v3.0/search/?q=*:*&rows=0 requested with an Accept: application/json header, reported 30,619,700 identifiers on 17 November 2026. ORCID’s own explanation of what the identifier is is worth two minutes if you do not have one yourself, because you will need one.
One discipline before you go further. A publication count is evidence about a record, not a verdict on a person. Plenty of excellent supervisors have modest counts because they teach heavily, work in a small field or publish long papers slowly. Use the numbers to ask better questions, never to rank human beings.
What does Indian regulation say about who can supervise you?
It says quite a lot, and almost none of it reaches applicants. The UGC Minimum Standards and Procedure for Award of Ph.D. Degree Regulations, 2022 set out who may be recognised as a Research Supervisor and how many scholars they may hold at once.
Recognition thresholds come first, in clause 6(1). A permanent Professor or Associate Professor needs a PhD and at least five research publications in peer-reviewed or refereed journals. A permanent Assistant Professor needs a PhD and at least three. A recognised supervisor cannot supervise scholars at another institution, where they may only act as co-supervisor, and adjunct faculty cannot be Research Supervisors at all. In fields with few or no refereed journals, the institution may relax the publication condition, with reasons recorded in writing.
Then come the caps, in clause 6(3): an eligible Professor, Associate Professor or Assistant Professor may guide up to eight, six and four PhD scholars respectively at any given time. Clause 7(1) allows up to two international scholars above that number, on a supernumerary basis, and clause 8 states that the total under a faculty member, as supervisor or co-supervisor, cannot exceed the clause 6(3) and 7(1) numbers together. Clause 6(5) adds the superannuation rule: faculty with less than three years of service left may not take new scholars, may continue with those already registered until superannuation, and may co-supervise afterwards but not past the age of 70.
The clause that changes your preparation, though, is clause 5(4), which requires the institution to maintain, on its own website:
a list of Ph.D. supervisors (specifying the name of the supervisor, his or her designation, and the department/school/centre), along with the details of Ph.D. scholars (specifying the name of the registered Ph.D. scholar, the topic of his/her research and the date of admission) admitted under them, updated every academic year.
Read that again as an applicant. The topics a supervisor actually takes, and how many scholars they are already carrying, are a record the institution is required to publish. You can read a supervisor’s current load, and the dates their students joined, before you write a single line of an email. Two practical readings follow. A cap is a ceiling, not a quota to fill, so a supervisor sitting at eight of eight is not accepting students this year whatever the department page implies. And an institution that does not publish the list at all has told you something about how it runs, which is itself a finding worth noting.
Coursework is also fixed rather than negotiable: clause 9(1) sets a minimum of 12 credits, including a research and publication ethics course and a research methodology course. Budget for it in your first-year plans.
Want the guided, hands-on version?
Our live Molecular Modeling & MD Simulations cohort bootcamp takes you from zero to running real docking and MD workflows, with a portfolio project for your grad-school applications.
Which public records can you pull, and what does each one not tell you?
Four records, each answering a different question, each with a blind spot. Pull all four before you decide, and treat the gaps as the reason to talk to people.
| Record | What it tells you | How current it is | What it cannot tell you |
|---|---|---|---|
| The institution’s clause 5(4) supervisor and scholar list | Who is recognised to supervise, their designation and department, and the name, topic and admission date of every scholar already under them | Required to be updated every academic year, so it can be up to a year behind | Whether a free slot is actually open to you, and anything at all about how the person supervises |
| Europe PMC author search | How many indexed papers carry that name, how many are recent, and where the person sits in the author order | Live, but reflects publication lag of a year or more | Which papers were really driven by the group, and which name belongs to which researcher when initials collide |
| ORCID record, or a national profile such as VIDWAN in India | A single identity with affiliations, works and funding as the researcher has chosen to claim them | Live, but only as complete as the researcher keeps it | Anything the researcher has not added, since the record is self-curated |
| Past students’ theses, including the acknowledgements | What the group was working on, who finished, how long they took and sometimes how the years felt | Historical by definition, often several years old | What the group is like now, especially after a move or a change in funding |
In India two directories make the third and fourth rows practical. VIDWAN, INFLIBNET’s expert directory, holds faculty profiles across Indian institutions, and Shodhganga holds Indian theses, where the acknowledgement pages are free and almost nobody reads them. IRINS, the research information system many Indian universities use for institutional profiles, is worth asking your library about if a department links to it.
What should you ask current students, and what does their silence mean?
Current students are the most direct evidence available to you, and the PLOS paper calls meeting them one of the most important steps before joining a lab. It also states the obvious problem: students who are unhappy with their current supervisor are unlikely to say so to a stranger who may become their labmate. The fix is to ask specific questions rather than general ones. Three of the paper’s own examples are worth asking verbatim, starting with “How often do you meet with your supervisor?”, then the typical turnaround time on a paper draft, then how they would describe the lab culture. It also suggests asking how the supervisor reacts to mistakes and to unexpected results, and how they handle interruptions from a student’s personal life.
Read the shape of the answer, not just its content. A specific number, a shrug and a story are three different answers. If every question returns a careful positive with no detail, you have learned that the student is being careful, which is information rather than reassurance. Ask to speak to two or three students in the group, not one, and ask the department office whether you can also meet students of other supervisors, which gives you a baseline for what normal looks like there.
What can past students and their theses tell you?
Past students answer the question current students cannot: what happened next. The PLOS paper recommends them for career outcomes and for a view of the whole process from start to finish, and notes that some of them no longer feel obliged to speak well of a former supervisor. It also points to a free source that costs nothing to read, which is the acknowledgement section of their theses.
Build a short list from the clause 5(4) record and from the group’s papers, then look for the names in a thesis repository. Two or three pages of acknowledgements will usually tell you who else in the lab mattered, whether the project changed shape, and how the years are described in the writer’s own words. Where the outcome you care about is leaving academia rather than staying, our guides to the skills a computational biology job actually asks for and moving into computational biology from a wet-lab background are a better map of where those alumni went.
Does career stage or group size matter more than the topic?
Both matter, in opposite directions, and the PLOS paper names them as the two structural attributes of a research group. Its own discussion of the trade-offs is summarised below. Nothing in this table is a rule about a particular person: it is a list of things to ask about.
| Attribute | What it tends to offer | What it tends to cost |
|---|---|---|
| Supervisor at a later career stage | Established research connections and settled protocols | Less room to shape the direction of the lab |
| Supervisor at an earlier career stage | More opportunity to shape the research direction, and a close view of how a lab gets built | Less access to academic political power, and less certainty about their supervision style, including to themselves |
| Large group | More collaborations, more learning from labmates, and strength in developing existing research topics further | Less supervisor time for each trainee |
| Small group | Better access to the supervisor, and room for more disruptive work | Can be more isolating |
Joining a brand new group is a real option and a real cost at once. The paper notes it is a good way to learn how to build a lab if that is your own ambition, while taking time and energy away from your thesis project.
What should you ask in the interview, and what are they asking you?
The PLOS paper frames the meeting as mutual: interview them as much as they interview you, prepare questions in advance and watch how they answer. Ask about lab culture, about where they want the research to go in the long term, and about what they look for in a graduate student. Then ask yourself whether you felt the need to put on an act beyond ordinary interview nerves, because representing yourself accurately is the only thing that makes the fit test meaningful. Discovering a poor fit in an interview costs you an afternoon. Discovering it after you commit costs years.
Prepare the practical side too. Our bioinformatics interview and coding test preparation guide covers what gets asked technically, and the bioinformatics CV guide covers the document that arrives with your email. If your record is still thin, the fastest honest fix is a finished, checkable piece of work rather than another line of coursework: start with our free certifications and assessments, then put the result where a supervisor can see it. The computational biology skills roadmap sets out the order to learn things in, and publishing your first paper as a student explains what the group’s output rate will mean for your own record.
What are the red flags, and how many supervisors should you approach?
The paper’s most useful section is also its least quantitative. It argues for treating every interaction, from the first email onward, as a piece of data, and then for listening to your own internal dialogue. Phrases like “it will be different for me”, or deciding that a student was probably exaggerating, are the warning signs. Genuine eagerness to start is the good one. Other red flags it names directly include an expectation that you are in the lab from early morning to midnight as a matter of routine, and an unwillingness to discuss project funding or stipend before you commit.
It also asks you to weigh the whole experience rather than the supervisor alone, including your mental health, pointing to published evidence that graduate students report disproportionately higher rates of depression and anxiety than the general population (Evans et al., Nature Biotechnology, 2018), and to survey evidence that many PhD students find their programme does not meet their original expectations (Woolston, Nature, 2019). Where you live, who is nearby and what you do outside the lab are part of the decision, not a distraction from it.
As for how many to approach, the paper’s last rule is to do the whole process again. Comparing options is what teaches you which red flags matter and what your own working style needs, and the authors reject the idea that you are wasting anyone’s time by looking carefully. Funded positions with named supervisors appear and close on their own schedule, so keep an eye on current research positions and fellowships while you compare, and read our own published work as a worked example of what a group’s record looks like from outside.
One honest caveat about the source. The PLOS paper was written by four PhD students drawing on their own supervisor searches, and it says so. That is why it is good on what to ask and deliberately quiet on how to find candidates in the first place, which it states plainly is outside its scope. The regulatory records above are the part it leaves to you.
Frequently asked questions
Should I email a supervisor before I apply?
For most routes, yes. Read two or three of their recent papers where they are senior author, then write a short message with one specific question that follows from the work. The cold-email mechanics are covered in our bioinformatics internship guide, and they transfer directly.
What if the only interested supervisor is already at the cap?
Then they cannot take you as supervisor this cycle. Clause 6(3) of the UGC regulations caps a Professor, Associate Professor and Assistant Professor at eight, six and four scholars at any given time, and clause 8 counts co-supervision towards that total. Ask when a slot is expected, and whether a co-supervision arrangement with a colleague is possible.
Does a co-supervisor count towards the limit?
Yes. Clause 8 states that the total number of PhD scholars under a faculty member, as supervisor or co-supervisor, cannot exceed the numbers in clauses 6(3) and 7(1) together. Co-supervisors from other departments or institutions may be appointed with the approval of the competent authority.
Should I choose the topic or the person?
The person. In the Indian system a Research Advisory Committee reviews your proposal and finalises the topic after you are admitted, so the topic is still movable while the supervisory relationship is not. Choose someone whose questions interest you and whose working style fits yours.
Is a supervisor with few publications a bad supervisor?
Not at all. Recognition thresholds set a floor of five publications for a Professor or Associate Professor and three for an Assistant Professor, and plenty of strong supervisors sit close to it because of teaching load or field size. Use the count to decide what to ask, not who to respect.
What if the institution has not published its supervisor list?
Clause 5(4) requires it, updated every academic year, so its absence is a reasonable thing to ask about. Write to the department or the PhD section and ask for the current list of recognised supervisors and scholars admitted under them. How that request is handled is itself useful information.
Written by the StemSkills Lab team, who have more than ten years of work in sequence and structural bioinformatics, drug discovery and design, and multiscale molecular modelling, and who supervise student research projects.
Want the guided, hands-on version?
Our live Molecular Modeling & MD Simulations cohort bootcamp takes you from zero to running real docking and MD workflows, with a portfolio project for your grad-school applications.
