Single Cell Sequencing and The Future of Drug Discovery

Drug discovery is a high-stake endeavor with a startling attrition rate in clinical trials due to unforeseen pharmacokinetics and toxicity issues.

Despite increasing efforts the number of successful drugs does not correlate with the massive research and development spendings.

Single cell RNA sequencing (scRNA-seq) is revolutionizing this landscape enabling researchers to dissect cellular mechanisms at unparalleled resolution.

By uncovering nuanced insights into drug targets, biomarkers, and patient responses, scRNA-seq can streamline drug development, and reduce costs by improving the success rates of clinical trials.

This accelerates the discovery of new therapeutics, and enhances the precision and efficacy of treatments, paving the way for a new era in personalized medicine.

Drug development is costly and time-consuming, taking approximately 10-15 years, and costing between $900 million to above $2 billion per drug.

At the root of such a high price tag is the repeated drug attrition – i.e. the high rate of failure during clinical development – that can be attributed to poor pharmacokinetics and toxicity causing failure at late stage. If drug developers have the right tools, these failures can be identified and addressed before they reach the clinical trial phase, potentially saving billions of dollars (Figure 1).

Figure 1: Sustained increases in pharmaceutical R&D spending do not necessarily lead to rising numbers of new drugs. R&D spending also reflects rising costs of researchers and laboratory technologies. Data extrapolated from the Congressional Budget Office for the year 2019.

In 2024, a team from the Wellcome Institute in Cambridge conducted a retrospective analysis of known drug target genes to identify potential predictors of clinical trial success using scRNA-seq data. They analyzed 30 diseases and 13 tissues to investigate potential predictors of target clinical success using scRNA-seq data from a publicly available database. The paper shows that drug targets have a cell type specific expression in disease-relevant tissues and is a robust predictor of a target’s progression from Phase I to Phase II clinical trials.

ScRNA-seq can help identify relevant cell types for a particular disease. This predictive power can streamline the drug development pipeline by focusing resources on the most promising targets.

Today’s drug discovery arena thrives on large volumes of data.

This is where scRNA-Seq comes in. Each scRNA-seq experiment is capable of generating millions to billions of datapoints.

This need becomes especially evident when analyzing a tissue at the single cell level. Sequencing with enough reads to capture rare cells or low-abundance transcripts — or using the technology capable of such detection regardless of the sequencing depth — generates millions of gene expression values.

Each cell operates as its own living entity, complete with organelles and unique functions, all interconnected through intricate pathways and networks. Each cell moves through life stages — resting, activation, proliferation, differentiation, senescence. This variety of single cells at different stages organize into tissues, which in turn make up organs, communicating and interacting as part of a larger community. When researchers compare different cell states, they encounter subtle but fundamental differences across samples. Such an enormous volume of data can train models to capture diversity within tissues, organs, individuals, or populations.

The high dimensional and high resolution information obtained through scRNA-seq data reveals detailed cellular characteristics not obtainable with bulk sequencing or low-throughput technologies.

These datasets are also invaluable for AI-driven drug development, which relies on large, high-quality scRNA-seq datasets to recognize patterns indicative of disease mechanisms or drug responses. As these AI models learn from expansive datasets, they become more adept at predicting outcomes, like which drugs are likely to succeed in clinical trials.

Using scRNA-seq researchers can assess the response of various cell populations in tissue samples to fine-tune drug dosage and enhance safety before clinical trials.

By predicting pharmacokinetics and potential toxicity early in the drug discovery phase, scRNA-seq can help filter out likely failures, reducing costs and improving success rates by minimizing investments in unsuccessful candidates. This approach is vital in addressing the high attrition rates often seen in clinical trials.

Target Identification and Validation

RNA sequencing at the single cell resolution is crucial for identifying genes linked to specific cell types or novel states involved in disease, aiding in the discovery of potential drug targets. For target validation, scRNA-seq supports gathering evidence across disease biology and target tractability. Study models, including cell lines and patient-derived organoids, can be evaluated to reveal cell-type-specific transcriptomic responses and pathway alterations related to disease states.

When scRNA-seq is used to analyze CRISPR perturbations, detecting the target genes and the cascade of pathway modifications triggered helps researchers understand complex interactions within cellular networks. This approach provides insights into gene function, regulatory mechanisms, and potential therapeutic targets.

Combining scRNA-seq with CRISPR screening allows for the large-scale mapping of how regulatory elements and transcription start sites impact gene expression in individual cells. This approach has been applied to profile approximately 250,000 primary CD4+ T cells, enabling systematic mapping of regulatory element-to-gene interactions and the functional interrogation of non-coding regulatory elements at the single-cell level.

Drug Screening

Traditional drug screening relies on general readouts like cell viability or marker expression, lacking comprehensive detail. ScRNA-seq enables detailed cell-type-specific gene expression profiles, essential for understanding drug mechanisms.
High-throughput screening now incorporates scRNA-seq for multi-dose, multiple experimental conditions, and perturbation analyses, providing richer data that support comprehensive insights into cellular responses, pathway dynamics, and potential therapeutic targets.

This approach enables researchers to identify subtle changes in gene expression and cellular heterogeneity, enhancing the understanding of drug efficacy and resistance mechanisms.

Biomarkers Identification and Patient Stratification

Biomarkers are characteristic features of a process that can be objectively and reproducibly assessed and measured. They can be prognostic, diagnostic, predictive, monitoring biomarkers.

Traditionally, biomarkers have been identified using several techniques. The advent of RNA sequencing has significantly accelerated this process, but bulk transcriptomics, while historically used to identify cancer biomarkers, fails to capture cell population complexity.

ScRNA-seq has advanced this field by defining more accurate biomarkers, such as those in colorectal cancer, leading to new classifications with subtypes distinguished by unique signaling pathways, mutation profiles, and transcriptional programs.

This deeper molecular understanding enables to evaluate the risk of developing a disease, monitor a disease course, and make an accurate diagnosis. It allows for more precise stratification of patients, tailored therapeutic strategies, and improved predictions of treatment responses, ultimately contributing to better clinical outcomes and personalized medicine approaches.

To fully harness the potential of big scRNA-seq data, selecting a platform that balances high-resolution output with scalability and flexibility can make the difference between uncovering actionable biological insights or missing critical data.

Parse Biosciences EvercodeTM v3 chemistry with combinatorial barcoding is a versatile method that can barcode up to 10 million cells in over a thousand samples in one experiment (Figure 2).

Figure 2: This first of its kind study measured 90 cytokine perturbations across 12 donors, 18 immune cell types, resulting in nearly 20,000 observed perturbations. This generated a 10 million cell dataset with 1,092 samples in a single run.

In this pioneering experiment, 90 cytokine perturbations across 18 immune cell types were measured in the peripheral blood mononuclear cells (PBMCs) from twelve donors, resulting in nearly 20,000 observed perturbations.

The experiment generated a 10 million cell dataset with 1,092 samples in a single run.

The study, the first of its kind, generated a large dataset that explored rare cell types that are hard to detect in smaller samples, using Bayesian models to make predictions and decisions based on the data. This study highlighted important considerations in large perturbation studies.

To illustrate the importance of analyzing large cell numbers to capture biological complexity, the authors downsampled a small PBMC subset —CD16 monocytes make up only 5-10% of the monocytes population. While cytokine effects were barely detectable in just 78 CD16 cells, increasing the sample size to 2,500 significantly boosted the number of differentially expressed genes.

Furthermore, the response of two small PBMC subsets—CD16 monocytes and CD4 memory T cells—to the cytokine IFN-omega revealed both shared patterns and unique cellular behaviors. While the two subsets displayed significant similarities in their overall response, individual cells exhibited distinct behaviors and reactions to IFN-ω. This complex and nuanced biological response would have gone unnoticed in a smaller dataset (Figure 3).

Figure 3: Heatmap showing expression of genes that are upregulated in response to IFN-ω. Subsets of genes here include genes that are upregulated in both CD4 Memory and CD16 Monocytes, CD4 Memory cells only, or CD16 Monocytes only.

This approach demonstrated that large screenings need large sample sizes to detect the behavior of all cells, including rare types and low-abundance transcripts. 

These datasets are also a gold mine for gathering robust data for training advanced models, which will enable scientists to make well-informed predictions and expedite their research progress.

The transformative power of scRNA-seq lies in its ability to generate high-dimensional datasets that reveal cellular heterogeneity, rare cell dynamics, and complex molecular interactions.

As these capabilities align with advancements in computational power and multi-omics integration, the future of drug discovery promises targeted therapies, precision medicine, and accelerated regulatory approvals.

With the right platforms, scRNA-seq is poised to drive monumental advances, making once-impossible breakthroughs within reach.