PacBio and TychoBio have announced a collaboration to generate full-length RNA data that TychoBio will use to train artificial intelligence (AI) models for RNA therapy design, as part of its work to identify new rare disease treatments.
PacBio, which is listed on Nasdaq under the ticker PACB, develops sequencing technologies. TychoBio is an AI drug discovery company. The announcement was made from Menlo Park, California, and Paris.
Sequencing data from more than 10,000 samples
Under the terms of the agreement, TychoBio plans to use PacBio HiFi sequencing and Kinnex long-read RNA sequencing to generate data from more than 10,000 samples.
The aim is to build large-scale datasets linking the design of steric blocking antisense oligonucleotides (SBOs) to their effects across the transcriptome. The companies said there are plans to expand the approach to siRNA and other RNA-targeting compounds.
Measuring changes across multiple cell types
TychoBio will test candidate RNA therapies and use PacBio sequencing to measure how each one changes the transcriptome across multiple cell types.
Those results will then become training data for TychoBio’s drug discovery models.
Predicting efficacy, duration and unintended effects
The models are designed to predict which compounds are most likely to work, how long their effects may last and where unintended effects may occur.
According to the companies, training models to maximise the on-target efficacy and effect duration of SBOs requires an in-depth understanding of the interaction between SBOs and the cell’s transcriptome.
By more fully characterising the effect of SBOs in various cell lineages, TychoBio expects to gain new insights into RNA regulation and the mechanism of action of SBOs. The companies said this work is intended to accelerate the development of safer and more efficient therapeutic designs.
Data quality behind the models
Felix Raimundo, founder and CEO of TychoBio, said the quality of the training data was central to the project.
“AI models are only as good as the data they are trained on. To effectively identify new rare disease treatments, we needed the highest quality biological data, which can only be generated using PacBio HiFi sequencing technology,” said Raimundo.
Raimundo added that HiFi long-read sequencing allows TychoBio to see across full-length transcripts. The collaboration will apply this approach first to SBOs before extending it to siRNA and other RNA-targeting compounds.
Frequently asked questions
What are PacBio and TychoBio collaborating on?
The companies are collaborating to generate full-length RNA data that TychoBio will use to train AI models for RNA therapy design.
How many samples will be sequenced in the PacBio and TychoBio collaboration?
TychoBio plans to use PacBio HiFi sequencing and Kinnex long-read RNA sequencing to generate data from more than 10,000 samples.
Which RNA therapies will TychoBio’s models focus on?
The datasets will link the design of steric blocking antisense oligonucleotides (SBOs) to their effects across the transcriptome. There are plans to expand the approach to siRNA and other RNA-targeting compounds.
What will TychoBio’s AI models predict?
The models are designed to predict which compounds are most likely to work, how long their effects may last and where unintended effects may occur.
Who leads TychoBio?
Felix Raimundo is the founder and CEO of TychoBio.
