Pipeline

Developing correctors for misfolding diseases requires a new approach to drug discovery

The correct Cellular Context

Predicting the structure of a protein is hard. Modeling the cellular journey of a misfolded protein, how the misfolding affects the protein’s function, and the cellular response to that stress, is even harder. We’ve developed a Generative Biology platform for studying protein misfolding in human cells, and to conduct high-throughput screening in that environment.

Multiplexed Read Outs

Engineering small molecule correctors requires developing a drug against multiple mutations from the start, while also optimizing the molecules for drug-like qualities. Octant’s Generative Biology platform enables the multiplexed measurement of the many biological phenomena associated with protein folding and trafficking across hundreds of mutations and conditions. This enables us to fine-tune our compounds to make effective new therapies for the most patients.

Proper folding and trafficking of the mutated protein triggers a DNA barcode identifying the cell line harboring that variant. Measuring barcodes using NGS enables multiplexed testing of many variants simultaneously.

Direct-to-biology HT-SAR

Correcting misfolded proteins is a molecular dance in a crowded cellular environment. It’s difficult to progress correctors of protein misfolding because traditional methods that rely on crystal structures and biochemical binding are often not amenable to unfolded proteins. Instead, the Octant Navigator weekly synthesizes and screens thousands of chemical analogs, iterating on a molecule’s efficacy and drug-like qualities. This combines the rationality of SAR with the serendipity of phenotypic screening to enable machine learning and accelerate the empirical discovery of new corrector therapies.

HT-SAR navigates to new regions of chemical space by iterating through thousands of chemical motifs that would be left unexplored in traditional med-chem campaigns.

Steps of Cell Signaling

Steps of Transcription

Other applications of mechanism based screening

Cellular mechanisms are fundamental to many complex problems in drug discovery. Using our genetic barcoding technology we model and screen against mechanisms ranging from transcription to the signaling pathways associated with GPCRs and kinases. The Octant Navigator is well-suited to high value targets from CNS to metabolism to immunology.

Navigating Retinitis Pigmentosa

Toxic accumulation of mistrafficked rhodopsin is rescuable by small molecule correctors, sometimes called chaperones

Drug discovery in engineered cell systems

RHO-associated autosomal dominant retinitis pigmentosa (RHO-adRP) is an inherited retinal disease affecting thousands of patients that currently has no disease-modifying treatment.

Octant is using the Navigator to design small molecule correctors that mitigate toxic accumulation of misfolded rhodopsin in the retina. We’ve engineered human cells to read out on trafficking and biophysical properties of RHO mutations, generating millions of data points from each screen that guide drug design.

Multiplexing to increase patient coverage

While most rare disease programs initially target a single mutation, our programs apply multi-mutant approaches that maximize the addressable patient population from the start. We engineer different cell lines for different mutations, each with an RNA barcode that identifies the variant. Using deep mutational scanning (DMS), Octant measured the efficacy and potency of OCNT-980 by building uniquely barcoded human cell lines for every possible single amino acid variant of RHO, probing the cell lines together under carefully controlled conditions.

Trafficking activity with no drug

Octant comprehensively classified RHO missense variants by building human cell lines for every possible variant to the RHO gene

Generative Chemistry applied to our RHO-adRP program evolves tool compounds into drug-like molecules with improved potency

Generative Chemistry Improves Potency and Drug-Like Properties

Nano-scale chemistry enables us to build thousands of analog molecules at a time and optimize them in learning loops. In the RHO-adRP program we iteratively built and screened 250,000+ molecules direct-to-biology, yielding thousands of structure-activity insights. We transformed early crude compounds with low potency and poor drug-like properties into potent, stable, orally available correctors that cross the blood-retinal barrier and engage the target in the retina to produce in vivo efficacy.

A new approach for correctors therapies

Our Pipeline