Context

Multiple drug discovery and academic groups have adopted image-based profiling in the last decade for many good reasons: akin to other profiling methods like transcriptomics or proteomics, it captures rich biological data that’s overlapping and complementary. But unlike those methods, it provides a more holistic view of cells that integrates multiple biological processes (genomics, proteomics, metabolomics), and at 10x the scale! (more on image-based profiling).

However, in an era of advanced computational methods and complex in vitro models like iPSC-derived models or 3D cultures, a single, static snapshot of fixed cells is no longer enough.

Industry leaders in drug discovery are now turning to the live-cell version of cell painting to better profile drugs in development.

Here are five of the key reasons why.

 

1. True physiological relevance

Using assays that better reflect the in-vivo environment enhances the odds of selecting the right drugs. Fixation and staining protocols, by their very nature, introduce stress and disrupt a cell's native state and architecture, which are at the center of what’s captured by the cell painting method. While cell painting captures a richness of data on cell response, it requires washing and fixation steps.

Figure 1: trade-offs of cell painting and brightfield imaging

In response, scientists have experimented with brightfield imaging for live cell profiling. Although cells remain unperturbed, a limited amount of biological information (mostly boiling down to cell count) can be captured with brightfield images. Likely less than a handful of groups, notably Recursion Pharmaceuticals, have developed enormous datasets backed by massive capital investments, and leveraging deep learning and vision transformer models, they have recently started using their gigantic database of cell painting images to predict staining on brightfield images. Both brightfield and advanced AI models like these are sensitive to artifacts in the light path that are ubiquitous in cell cultures, bringing bias in the data that can make brightfield imaging very difficult to work with.

Live cell painting comes handy when scientists prefer assaying cells in their live, native environment, or when they are not ready just yet to bet on brightfield imaging which likely comes with hundreds of millions of dollars and waiting multiple years before even starting their assays.

ChromaLIVE, the only non-toxic live cell painting dye, allows to capture rich biological data, like cell painting, but without perturbing cells. It is the only biologically inert dye with hard data, such as gene expression data, supporting this description. As shown in multiple studies, cells cultured with ChromaLIVE for weeks remain healthy and unperturbed. This means the profiling data observed with ChromaLIVE are a more direct result of your drug candidate, and not a reaction to the assay conditions.

Figure 2: ChromaLIVE data on biological inertness. Link to publication.

Ultimately, this approach ensures a higher fidelity readout, leading to more reliable predictions.

 

2. Enhanced profiling insights for better decisions

To make the best possible decisions in drug discovery, the goal is to design screening assays that are sensitive enough to catch the most promising candidates while also being specific enough to weed out unviable compounds as early as possible. In other words, you want to minimize false negatives and also minimize false positives.

Live cell painting with ChromaLIVE is showing remarkable performance, which makes it a strong contender as the leading technology for image-based profiling:

• Improving bioactivity detection: in reference pharma customer datasets that used compounds known to be bioactive through a combination of dozens of assay technologies performed over multiple years, ChromaLIVE shows increased bioactivity detection when compared to cell painting at a fixed timepoint. By allowing the acquisition of multiple timepoints, ChromaLIVE is able to detect broader windows of bioactivity: early, late and even transient.
• Optimal balance between false positives and false negatives: the average precision score (mAP) for ChromaLIVE on reference cell-painting compounds is overall similar to -- and sometimes higher than -- cell painting. The mAP is a statistical framework developed for high-content profiling assays, and it essentially integrates both recall and precision, which can respectively be thought of as measures of false negative and false positive rates.

Figure 3: Improved bioactivity detection from ChromaLIVE

Figure 4: Comparing mAP of ChromaLIVE (y axis) and cell painting (x-axis) on a reference set of 90  compounds. Link to Broad's SBi2 2023 poster. More in the 2024 Broad pre-print here.

Not only can this approach yield similar or better results than traditional cell painting, but it can also help drug discovery leaders make the right calls on experimental drugs

3. Effortless workflows and integration across drug discovery

Traditional cell painting is a multi-step, labor-intensive process involving fixation and multiple wash steps. This complexity is a bottleneck for high-throughput screening and validation.

Live cell painting with ChromaLIVE is designed for efficiency and automation. It is a one-step, mix-and-read dye that requires zero wash steps. This not only saves significant time and resources but also drastically reduces the risk of cell loss and experimental variability. This streamlined workflow makes it simple to run multiple time points in a single campaign.

It also makes for an ideal tool that can be used across discovery pipeline stages. For instance, drug discovery scientists have used ChromaLIVE in the assay validation phases to assess multiple timepoints at once, and again later in target ID or drug screens, allowing to keep the same methodology for minimal protocol adaptations and variability throughout the discovery process.

Figure 5: ChromaLIVE protocol workflow

4. Compatibility with complex and sensitive in vitro models

The most physiologically relevant in-vitro models—like patient-derived cells, iPSCs, neurons, and 3D organoids— also tend to be the most sensitive. Their delicate architecture and biology are often incompatible with commercial dyes marketed as “live-cell dyes”.

ChromaLIVE's gentle, non-toxic nature makes it the ideal tool for these systems. It provides uniform staining of 3D cultures without damaging their intricate structure, opening the door to effective profiling of 3D cell response - i.e. 3D cell painting. It can also track neurons or stem cells over long periods of time and measure their phenotypic changes.

This helps drug discovery teams more easily integrate predictive disease models earlier in the drug discovery pipeline.

Figure 6: Live Hek293 spheroids (not fixed, no clearing agent).

 

5. Finding new drug insights in kinetics

While fixed-cell assays are valuable, they only show a cell at one specific moment. The real biological story often lies in the "how" and "when" of a cellular response.

Live cell painting with ChromaLIVE allows the capture of new kinetic insights and to detect transient phenotypes—those subtle, fleeting changes that would be missed by a fixed-point assay. As highlighted in our recent data, ChromaLIVE can reveal the time-to-onset of a drug's effect, allowing for the discrimination of fast-acting from slow-acting compounds. This dynamic view provides biological information that simply isn’t mapped by conventional methods and can lead to the discovery of more potent or effective drug candidates.

Figure 7: Transient phenotype highlighted by ChromaLIVE from 12h-24h, with a different terminal phenotype afterwards.

In summary, the non-toxic, live-cell approach to drug profiling provides a richness of data that is otherwise invisible with fixed cell approaches.

 

Want to learn more about how ChromaLIVE can transform your screening workflow?

Learn More or See Our Protocol Here !