424 people
Across three lung-model cohorts
Including an independent evaluation at Tata Memorial Hospital.
Read the evidence summary BIOMARKERS × MACHINE LEARNINGEarlier
Earlier
changes
everything.
We bring blood biomarkers and machine learning together to look for cancer earlier. VarSee Lung is our first product, available for research use.
Can I take the test? Ask about availabilityScroll to follow the signal PUNE, INDIA · A GLOBAL AMBITION
New investors · October 2026
A warm welcome to
Anand Deshpande andSonali Anand Deshpande
They join us as investors, backing our work on turning blood biomarkers into actionable cancer intelligence using advanced machine learning.
Read the announcement
Supported by
RESEARCH PROGRESS
From the lab to independent evaluation.
3 hospitals
Multicentre study ethics approval
For a VarSee Lung diagnostic accuracy study at Tata Memorial Centre sites.
See the study collaborationsLung → breast
The next disease programme
Breast research with Sylvester Comprehensive Cancer Center, University of Miami.
Explore the development programmesFollow the signal
THE BIOMARKIQ APPROACH01 / 03
Biological signals
A signal is a start.
Blood carries proteins that change when a cancer starts to grow. One biomarker gives one view of that change.
Start with the biologyA panel sees more.
Cancer leaves more than one kind of clue. A panel measures several biomarkers at once, so the pattern between them can be read.
Explore the panel approachSpecialists connect it.
Specialist models each read the panel for one cancer. A gating model weighs them, with the person’s context, into one result.
Meet the MoE engine02WHY A PANEL?
One marker.
Part of the story.
Some biomarkers catch more cancers. Others clear more people without cancer. Watch what a second signal reveals in the same group of samples.
Why complementarity mattersHow a signal is judged
Find people with the condition.
Of the people with the condition, how many does the test find? Everyone past the threshold is flagged; squares short of it are missed.
Identify people without it.
Of the people without the condition, how many does the test correctly clear? Circles past the threshold become false alarms.
Add a second signal.
A second measurement can pull the groups apart, so fewer are missed and fewer are flagged by mistake. Whether a real panel does this is what validation has to show.
03THE MOE ENGINE
Different experts.
Read together.
Each specialist model reads the panel for one cancer. A gating layer weighs their answers with the person’s context, so one result reflects the whole pattern.
- InputsProtein panelClinical contextAge, sex, smoking, recent blood count
- Specialists
- Lung
- Breast
- Liver
- Colorectal
- Healthy baseline
- GateWeighs every reading with the person’s context
- One resultFor example: signal, likely lungor indeterminate, with a next step
One blood draw, many measurements.
A panel of proteins is measured from a single sample, alongside routine clinical context: age, sex, smoking and a recent blood count.
A specialist for each cancer.
Each specialist model is trained on the proteins relevant to one organ, and a healthy-baseline specialist reads what normal looks like. Each returns its own reading.
A gate weighs them, in context.
A gating model weighs every specialist’s reading together with the person’s clinical context, so no single signal decides on its own.
One result, or a next step.
The output is one result with its likely tissue of origin. When the readings disagree, it is indeterminate, with a directed next step instead of a forced call.
04ONE DISEASE AT A TIME
Lung first.
More cancers to follow.
Multi-cancer detection is the goal. We get there one cancer at a time, and lung comes first.
MEET OUR FIRST PRODUCT
VarSee Lung
A blood panel and a specialist model built around the biology of lung cancer. A multicentre study has ethics approval at three Tata Memorial Centre hospitals.
For research use onlyExplore VarSee Lung
Life leaves a pattern.
We’re learning to read it earlier.
RESEARCH COLLABORATIONS
A shared question.
Across centres.
Our collaborations with Tata Memorial Centre institutions for the VarSee Lung Diagnostic Accuracy Study.
Tata Memorial Centre(opens in a new tab)Mumbai
Varanasi
Sangrur
Study ethics clearance announced .
Read the study announcement (LinkedIn, opens in a new tab)05WHAT WE OPTIMIZE FOR
Finding cancer matters.
False alarms matter too.
Detecting more cancers matters. So does avoiding unnecessary alarms. We study both, disease by disease.
FROM THE BIOMARKIQ LAB06THE PEOPLE DOING THE WORK
Cancer biology.
Machine learning.
One team.
Manasi, Snahlata, and Atharva founded BioMarkIQ to bring cancer biology and computational thinking into the same lab.
Today a team of ten works on the panels, the models and the studies.
Meet the team
Interested in the test?
Help build what comes next?
Ask about VarSee Lung, research partnerships, or investing in BioMarkIQ.