NewLimit was founded to significantly extend human healthspan. We are developing epigentic reprogramming medicines to treat age-related diseases.
Our science draws upon remarkable work done by the broader biotechnology community. Below, we outline some of the prior research that inspires our approach.
Epigenetic reprogramming to address age-related disease
Researchers have reported that epigenetic reprogramming can reduce the burden of age-related disease across diverse pre-clinical models. These results are still early, and we view them as hints that reprogramming can provide benefit in each context, rather than definitive evidence. There is still a large amount of discovery science to be done in each indication.
Disease Model | Intervention | Reference |
---|---|---|
Type I diabetes | Partial OSKM, transgenic | https://pubmed.ncbi.nlm.nih.gov/27984723/ |
Skeletal muscle injury | Partial OSKM, transgenic | https://pubmed.ncbi.nlm.nih.gov/27984723/ |
Skeletal muscle injury | OSKMLN, mRNA + LNPs | https://pubmed.ncbi.nlm.nih.gov/32210226/ |
Skeletal muscle injury | OSKM, transgenic | https://pubmed.ncbi.nlm.nih.gov/34035273/ |
Glaucoma | Partial OSK, AAV | https://pubmed.ncbi.nlm.nih.gov/33268865/ |
Myocardial infarction | Partial OSKM, transgenic | https://pubmed.ncbi.nlm.nih.gov/34554778/ |
Skin wound healing | Partial OSKM, transgenic | https://pubmed.ncbi.nlm.nih.gov/29761584/ |
Skin wound healing | Partial OSKM, long-term, transgenic | https://www.nature.com/articles/s43587-022-00183-2#Sec28 |
Liver regeneration | Partial OSKM, transgenic | https://pubmed.ncbi.nlm.nih.gov/35476977/ |
Pooled screening to search through large, combinatorial hypothesis spaces
Traditionally, reprogramming hypotheses have been tested in low throughput, using low-dimensional read-outs of cell state. Single cell and functional genomics technologies have unlocked a higher throughput approach.
NewLimit’s discovery engine is built upon the shoulders of these technological innovations, allowing us to run experiments at a scale that is orders of magnitude larger than traditional approaches.
Machine learning models enable in silico experiments
Machine learning models now enable us to learn the rules of biological systems from data. We can use these models to perform “in silico experiments” — predicting the outcome of an experiment that has not yet occurred using the model’s learned representation of biology.
These tools allow us to run experiments in the world of bits so that we can prioritize the experiments we run in the world of atoms. We’ve been inspired by performance of in silico experiments on diverse biological tasks related to our key questions.
Task | Reference |
---|---|
Predict diverse cell structures from a simple brightfield image | https://www.nature.com/articles/s41592-018-0111-2 |
Predict sequence function and mutation effects from DNA sequence alone | https://www.nature.com/articles/s41592-021-01252-x |
Predict the effect of gene knockdowns on gene expression | https://www.science.org/doi/abs/10.1126/science.aax4438 |
Predict chromatin accessibility from mRNA expression and vice-versa | https://www.pnas.org/doi/10.1073/pnas.2023070118 |
Immunology
We are working to restore function in aged T cells as one of the earliest applications of our technology. T cell aging and dysfunction is linked to a variety of pathologies and limits the performance of cell therapies. Emerging evidence suggests that there is an epigenetic basis for the dysfunction that arises with both age and disease, and that this dysfunction can be reversed by reprogramming.
Description | Reference |
---|---|
T cells enter a common dysfunctional state with age | https://pubmed.ncbi.nlm.nih.gov/33271118/ |
T cell exhaustion is a pathological cell state with an epigenetic basis | https://pubmed.ncbi.nlm.nih.gov/27789799/ |
Reprogramming can restore function in dysfunctional T cells | https://pubmed.ncbi.nlm.nih.gov/23290140/ |
https://www.nature.com/articles/s41590-021-00964-8 |