Our science

Built on a decade of foundational research

Reflector Bio's platform emerges from years of foundational work in machine learning for single-cell biology and perturbation modeling. Our team has published in the field's leading journals, establishing the methods that now power our platform.

Featured

Scalable and universal prediction of cellular phenotypes

Ji Y, Tejada-Lapuerta A, Schmacke NA, Zheng Z, Zhang X, Khan S, Rothenaigner I, Tschuck J, Hadian K, Hornung V, Theis FJ

We reduce experimental screening volume by orders of magnitude while providing mechanistic insight. Our model integrates diverse functional readouts across assays and cell types into a unified predictive architecture.

Selected publications

From our team

Nature Methods
2025

Pertpy: an end-to-end framework for perturbation analysis

We built an open-source toolkit that enables reproducible analysis workflows across large-scale screens, systematically quantifying and annotating small molecule and genetic effects.

Nature Biotech
2025

Defining and benchmarking open problems in single-cell analysis

We established community standards for comparing computational methods through a benchmarking platform that evaluates consistency across core single-cell tasks.

Nature Medicine
2023

An integrated cell atlas of the lung in health and disease

We contributed to the Human Cell Lung Atlas, showing how integrated data at scale reveals disease-relevant cell states and new sources of biological variation.

Cell Systems
2021

Machine learning for perturbational single-cell omics

We established the foundation for learning cellular response from large-scale experimental data, defining the key modeling challenges and opportunities that now guide our platform.

Pertpy: an end-to-end framework for perturbation analysis

We built an open-source toolkit that enables reproducible analysis workflows across large-scale screens, systematically quantifying and annotating small molecule and genetic effects.

Defining and benchmarking open problems in single-cell analysis

We established community standards for comparing computational methods through a benchmarking platform that evaluates consistency across core single-cell tasks.

An integrated cell atlas of the lung in health and disease

We contributed to the Human Cell Lung Atlas, showing how integrated data at scale reveals disease-relevant cell states and new sources of biological variation.

Machine learning for perturbational single-cell omics

We established the foundation for learning cellular response from large-scale experimental data, defining the key modeling challenges and opportunities that now guide our platform.

Open science

Advancing the field through open-source contributions

Our commercial platform builds on research we have contributed to the open-source community. We believe that foundational tools should be accessible to all.

GitHub