Jian Shu, PhD, an investigator in the Cutaneous Biology Research Center at Massachusetts General Hospital and an assistant professor of Dermatology at Harvard Medical School as well as an associate member of Broad Institute of MIT and Harvard, is co-senior author of a new paper in Nature Biotechnology, Prediction of Single-Cell RNA Expression Profiles in Live Cells by Raman Microscopy with Raman2RNA

What Question Were You Investigating with this Study?

Cells are basic units of life. Decoding the genomics information of cells can uncover cell states — whether healthy or diseased states. AlphaFold from Google’s DeepMind revolutionized protein structure predictions through artificial intelligence (AI).

Moving beyond protein structure prediction, can we develop a technology that allows us to query and predict single-cell genomics information from images using AI at low cost, non-destructively and at large scale, with the potential to apply to humans in vivo?

Single-cell RNA-seq (scRNA-seq) and other single-cell profiling assays have opened new windows into understanding the properties, regulation, dynamics, and function of cells at unprecedented resolution and scale.

However, these assays are inherently destructive, precluding us from tracking the temporal dynamics of live cells, in culture or organisms.

Conversely, Raman microscopy offers a unique opportunity to collectively report on the vibrational energy levels of molecules in a label-free and non-destructive manner at a subcellular spatial resolution, but it lacks specificity and direct molecular information.

What Approach Did You Use?

We developed Raman2RNA (R2R), an experimental and computational framework to infer single-cell expression profiles in live cells through label-free hyperspectral Raman microscopy images and multi-modal data integration and domain translation.

We used spatially resolved single-molecule RNA-FISH as anchors to link scRNA-seq to the paired spatial hyperspectral Raman images, and trained machine learning models to infer expression profiles from Raman spectra at the single-cell level.

What Were the Results?

We were able to predict scRNA-seq profiles non-destructively from Raman images using either anchor-based integration with smFISH, or anchor-free generation with adversarial autoencoders.

R2R outperformed inference from brightfield images (cosine similarities: R2R>0.85 and brightfield<0.15).

In reprogramming of mouse fibroblasts into induced pluripotent stem cells, R2R inferred the expression profiles of various cell states.

With live-cell tracking of mouse embryonic stem cells differentiation, R2R traced the early emergence of lineage divergence and differentiation trajectories, overcoming discontinuities in expression space.

R2R lays a foundation for future exploration of live genomic dynamics.

What are the Implications?

This paper proposed a paradigm-shift concept in decoding cell states through AI and demonstrated for the first time how to predict single cell gene expression from images through AI, which will lay the foundation of non-invasive diagnosis, massively scalable drug discovery and change how we measure cell states in research and clinics.

The paper’s other senior authors are Tommaso Biancalani, a former Broad Institute scientist; and Aviv Regev, executive vice president at Genentech Research and Early Development, who is on leave from faculty positions at the Broad Institute and MIT’s Department of Biology.

Paper Cited:

Kobayashi-Kirschvink, K.J., Comiter, C.S., Gaddam, S. et al. Prediction of single-cell RNA expression profiles in live cells by Raman microscopy with Raman2RNA. Nat Biotechnol (2024). https://doi.org/10.1038/s41587-023-02082-2

About the Massachusetts General Hospital

Massachusetts General Hospital, founded in 1811, is the original and largest teaching hospital of Harvard Medical School. The Mass General Research Institute conducts the largest hospital-based research program in the nation, with annual research operations of more than $1 billion and comprises more than 9,500 researchers working across more than 30 institutes, centers and departments. Mass General is a founding member of the Mass General Brigham healthcare system.