APPLICATIONS OF PHENOTYPIC SCREENING FOR TARGET IDENTIFICATION IN DRUG DISCOVERY
Pooled CRISPR screens are widely used to supercharge drug target identification and validation. At Myllia, we perform CRISPR-based functional genomics screens in primary human cells that capture physiologically relevant human disease biology. We enable best-in-class target discovery at scale by leveraging high-content phenotypic read-outs such as fitness, flow cytometry, CITE-Seq and single-cell transcriptomics (CROP-Seq) - paving the way for next-generation drug targets addressing unmet medical needs in immunology and oncology.
Drug target ID/validation incl. deconvolution and identification of disease-associated gene function using functional genomics
Drug target ID/validation incl. deconvolution and identification of disease-associated gene function using functional genomics
Target identification is the first step of a drug discovery campaign and begins with high-content screening to identify possible ‘druggable’ targets and their role in the respective disease. Myllia’s unique CROP-Seq (Perturb-Seq) technology for phenotypic screening in combination with the best available cellular models (incl. cancer cells and primary human cells) supports the identification and validation of critical genes and pathways driving certain disease states.
Genome-wide association studies (GWAS) contribute to target identification in drug discovery, identifying thousands of genetic variants that are linked to disease. Unfortunately, many of these loci lie in non-coding regions of the genome. Pinpointing the gene(s) whose expression is regulated by these regions would elucidate novel drug targets that are causally linked to disease. Myllia has built a high-content screening platform based on CRISPR interference (CRISPRi) that can map disease-associated variants to genes in an unbiased fashion.
Mode of action (MoA) analysis and high-content screening for drug signatures
Mode of action (MoA) analysis and high-content screening for drug signatures
Understanding how drugs act in the complex environment of a cell remains one of the critical aspects of drug discovery and development after initial drug target identification. Importantly, phenotypic screening with CROP-Seq (Perturb-Seq) delivers transcriptional profiles associated with drug action and indicates which genes impact the drug profile, thus providing unique insights into its mechanism of action. It also uncovers genes that modify drug responses, thus paving the way for combination therapy approaches.
CROP-Seq perturbation datasets for training of AI/ML-based foundation models of human cell biology
CROP-Seq perturbation datasets for training of AI/ML-based foundation models of human cell biology
A foundation model of the human cell is a digital representation of the biology of a human cell. While we see a plethora of applications for such a model, we would like to highlight three applications in particular:
1. A foundational model of human cells could learn perturbation outcomes from simplistic models and “transfer” these into a novel “context”, i.e., a more complex cellular model. This would allow researchers to refine their hypothesis regarding perturbation outcomes and apply them to different biological contexts, thus yielding hypothesis that have a greater predictivity.
2. Many diseases may require the inactivation of more than one drug target. However, even the space of all possible gene pairs is way too large (~200 million gene pairs) to be experimentally addressed. This requires computational tools that predict the gene pairs enabling new combination therapies for diseases of unmet need.
3. Imputation – A computational approach that allows the inference of the complete transcriptome from a limited snapshot. Specifically, this could reduce the NGS requirements of these experiments, thus making large-scale endeavours more cost-effective and feasible in future.
Pooled CRISPR screens in human-centric, primary immune models, e.g., primary helper T cells and myeloid cells
Pooled CRISPR screens in human-centric, primary immune models, e.g., primary helper T cells and myeloid cells
Engineering of T-lymphocytes and myeloid cells has become a crucial factor driving the development of novel cellular medicines and cancer immunotherapies. However, apart from tumor-associated antigens (TAA) and Chimeric Antigen Receptor (CAR) or T Cell Receptor (TCR) discovery, many immune cell-intrinsic features involved in primary immune cell potency remain elusive. At Myllia, CRISPR/Cas9-based high-content screens are performed in primary human cells, e.g., to study primary T cell activation, differentiation and phenotypic plasticity. Utilizing dropout fitness/viability screens, FACS-based screens or CROP-Seq screens at single-cell resolution, we aim to partner with pharmaceutical companies striving to develop the next-generation of primary immune cell therapeutics.