Interrogatory cell-based assays and uses thereof
Inventors
VISHNUDAS, VIVEK K. • Sarangarajan, Rangaprasad • Narain, Niven Rajin • Du, Min • Walshe, Tony
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12437835-B2
Publication Date
2025-10-07
Expiration Date
Abstract
Methods for identifying a modulator of angiogenesis and methods for modulating angiogenesis in a mammalian subject are described herein. In some embodiments, the methods include obtaining data sets from a model for angiogenesis and generating a causal relationship network model based on the obtained data. In some embodiments, the methods include identifying, from the causal relationship network model, a causal relationship unique in angiogenesis, where a gene, lipid, protein, metabolite, transcript, or SNP associated with the unique causal relationship is identified as a modulator of angiogenesis.
Core Innovation
The invention provides an interrogative, AI-driven platform for identifying a modulator of angiogenesis by integrating multi-modal cellular characterization with functional cellular response data. The method obtains a first data set characterizing cells associated with angiogenesis using genomic data, lipidomic data, proteomic data, metabolomic data, transcriptomic data, and single nucleotide polymorphism data, and a second data set representing functional activities or cellular responses of the same cells associated with angiogenesis.
A first causal relationship network model is generated on a programmed computing system including a plurality of processors. The generation includes creating a list of network fragments, determining probabilistic scores for the fragments based on the first data set and the second data set, creating an ensemble of trial networks from different subsets of the network fragments, and globally optimizing the ensemble by evolving trial networks in parallel. Relationships and causality in the first causal relationship network model are determined based on the first data set and the second data set and not based on previously identified or known biological relationships between variables.
A differential causal relationship network is then generated from the first causal relationship network model and a second causal relationship network model based on control cell data. The differential causal relationship network is formed by including relationships absent from the other model and including relationships with at least one significantly different parameter in the other model. From this differential causal relationship network, a causal relationship unique in angiogenesis is identified, and a gene, lipid, protein, metabolite, transcript, or SNP associated with the unique causal relationship is identified as a modulator of angiogenesis.
Claims Coverage
Two independent methods are provided. The coverage centers on causal-network modeling from multi-omics and functional or cellular-response data, using network fragments, probabilistic scoring, and parallel global optimization, and on differential causal relationship network construction against control cell data to identify angiogenesis-unique causal relationships. The claims exclude previously identified or known biological relationships as a basis for causality determination.
Causal relationship network from multi-omics and functional responses for angiogenesis
Generating a first causal relationship network model among genomic data, lipidomic data, proteomic data, metabolomic data, transcriptomic data, and single nucleotide polymorphism data characterizing cells associated with angiogenesis, and one or more functional activities or cellular responses of the cells, using a programmed computing system including a plurality of processors.
Network-fragment ensemble probabilistic scoring with parallel global optimization
Creating a list of network fragments, determining probabilistic scores for each fragment based on the first data set and the second data set, creating an ensemble of trial networks using different subsets of the fragments, and globally optimizing the ensemble by evolving the trial networks in parallel using the plurality of processors.
Causality determined without previously known biological relationships
Determining relationships and causality in the first causal relationship network model based on the first data set and the second data set and not based on previously identified or known biological relationships between variables.
Differential causal network versus control cell model using significantly different relationship parameters
Generating a differential causal relationship network from the first causal relationship network model and a second causal relationship network model based on control cell data, including relationships absent from the other model and relationships with at least one significantly different parameter in the other model.
Identify angiogenesis-unique causal relationships and corresponding modulators
Identifying, from the differential causal relationship network, a causal relationship unique in angiogenesis, and identifying a gene, lipid, protein, metabolite, transcript, or SNP associated with the unique causal relationship as a modulator of angiogenesis.
The claims are directed to constructing a causal relationship network from multi-omics/genetic characterization and functional cellular responses, using an ensemble network-fragment approach with probabilistic scoring and parallel global optimization, then forming a differential causal relationship network versus control cell data to isolate causal relationships unique to angiogenesis and identify associated genes, lipids, proteins, metabolites, transcripts, or SNPs as modulators.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?