research focuses
Original research doesn't have to wait for graduate school.
Our four departments
We focus on the fields where we believe student researchers can do rigorous work now, and where that work is most likely to compound: life sciences, computation, mathematics, and AI for science.
Scientific training has long assumed that original research begins in graduate school. As the tools of research become cheaper and more accessible, that assumption looks less like a fact about talent and more like a bottleneck we have chosen to keep. Nishan was built to remove it. Our work is organized into four departments: Life Sciences, Computational Sciences, Mathematical Sciences, and AI for Science. Each is a field where a well-mentored student can produce work of real scientific value, but where the opportunity to do so is unlikely to reach most students if left to the existing pipeline. So we work alongside the people who know these fields best, including faculty, graduate researchers, and clinicians, to make sure the work happens sooner, and for more students, than it otherwise would.
Students accepted to a team receive mentorship from active researchers, a defined role on a project with a real question behind it, and the tools to carry it out, at no cost. We also build infrastructure the whole field can use: methodological guides and open dataset libraries, IRB navigation resources, author guidelines and a manuscript template, and the AI research tooling in our Helix Lab, which we publish alongside documentation of its limitations.
Already, seven teams and more than 28 students are at work across the departments. One is examining how postpartum endocrine transitions interact with the cognitive-emotional processes underlying maternal mental health. Another is testing whether topological methods detect structural change in dynamic networks that conventional graph statistics miss. A third is asking whether epigenetic age acceleration predicts immune decline across the human lifespan. The first results are weeks away. We are committed to sharing candidly what they show, including where we were wrong.
Within Nishan's four research departments, Life Sciences is the largest, and it is where the access problem is sharpest. Biological research is expensive, slow, and heavily gated by institutional affiliation, so the questions a student can realistically work on are usually determined by which building they happen to have keys to. Wet lab access, IRB approval, licensed datasets, and a mentor with time are not distributed according to who is capable of using them. Our teams work on molecular biology, chemistry, genetics, neuroscience, pharmacology, and biomedical engineering, and they are designed around questions that can be answered rigorously with computational methods, public cohort data, and structured literature synthesis. That constraint is not a compromise. Much of the most useful work in modern biology now happens on data that already exists and has not been examined carefully enough.
Some of what our teams pursue is well-posed enough to close in a single cohort. Other questions are not, and those are the ones worth putting students on early. One team is examining how postpartum endocrine transitions interact with the cognitive-emotional processes underlying maternal mental health, a relationship that decades of depression and anxiety research have left underspecified. Another is asking whether dormancy mechanisms that evolved independently across nearly every branch of life share conserved molecular architecture. A third is mapping how TGF-β1 gradients shape tumor-associated macrophage polarization in pancreatic ductal adenocarcinoma. A fourth is testing whether epigenetic age acceleration predicts measurable immune decline in population cohorts with paired methylation and immunological data.
Doing this work properly requires infrastructure that most students never see. We build and publish it: methodological guides written with faculty partners, curated open-access dataset libraries, IRB navigation resources, statistical analysis frameworks, and annotated literature review templates. Teams that need computational support draw on our Helix Lab, where AI research tooling is paired with instruction in how those systems reason and where they fail, because a student who cannot evaluate a model's output should not be relying on one. Finished work goes to our Journal of Youth in Medical Science under double-blind review by faculty-affiliated evaluators, against criteria we publish openly. We are pursuing indexing in PubMed and Web of Science.
None of this is useful if it only reaches students who were already going to find their way into a lab. Our outreach programs bring research methodology training and mentorship directly to under-resourced schools, rural districts, and community colleges, and we measure them by whether participants go on to conduct original research and persist in their fields, not by attendance. Applications to research teams are open to any student, at any institution, at no cost. The first results from our current cohort arrive in the coming weeks, and we will publish what they show.