US 8,478,535 B2 · Systems and methods that utilize machine learning algorithms to facilitate assembly of aids vaccine cocktails

Abstract

The subject invention provides systems and methods that facilitate AIDS vaccine cocktail assembly via machine learning algorithms such as a cost function, a greedy algorithm, an expectation-maximization (EM) algorithm, etc. Such assembly can be utilized to generate vaccine cocktails for species of pathogens that evolve quickly under immune pressure of the host. For example, the systems and methods of the subject invention can be utilized to facilitate design of T cell vaccines for pathogens such HIV. In addition, the systems and methods of the subject invention can be utilized in connection with other applications, such as, for example, sequence alignment, motif discovery, classification, and recombination hot spot detection. The novel techniques described herein can provide for improvements over traditional approaches to designing vaccines by constructing vaccine cocktails with higher epitope coverage, for example, in comparison with cocktails of consensi, tree nodes and random strains from data.

Claims47

Dependent claims

A system that facilitates determining an epitome that provides a basis for a vaccine cocktail, comprising:

a processing unit;
a system memory coupled to the processing unit;
an input component that receives a plurality of overlapping patches xS corresponding to a set of subsequences from one or more pathogen sequences in a population;
and
a modeling engine that employs one or more machine learning algorithms to determine an epitome based on the plurality of overlapping patches xS, the epitome providing the basis for the vaccine cocktail, wherein the one or more machine learning algorithms comprises an expectation-maximization (EM) algorithm that includes an initial random guess for the epitome and iteratively reduces a free energy of the epitome converging to a local minimum of free energy, wherein the free energy combines T cell binding energy and HLA binding via a variational mapping distribution of respective ones of the plurality of overlapping patches xS to the epitome.

2 more elements

A system that facilitates identifying an epitome for generating AIDS vaccine cocktails, comprising:

a processing unit;
a system memory coupled to the processing unit;
an input component configured to receive a plurality of overlapping patches having sequences and store at least a subset of the plurality of overlapping patches in the system memory;
and
a machine learning component configured to employ machine learning to model sequence diversity for identifying an epitome to facilitate AIDS vaccine cocktail assembly, wherein the machine learning component comprises:
a cost function that accounts for acts that are needed to mount an effective immune response;
and
an expectation-maximization (EM) algorithm that includes an initial random guess for the epitome and iteratively reduces a free energy of the epitome converging to a local minimum of free energy, wherein the free energy combines T cell binding energy and HLA binding via a variational mapping distribution of respective ones of the plurality of overlapping patches to the epitome.

5 more elements

Dependencies are inferred from the claim text.

Systems and methods that utilize machine learning algorithms to facilitate assembly of aids vaccine cocktails

Life of this patent

Systems and methods that utilize machine learning algorithms to facilitate assembly of aids vaccine cocktails

Life of this patent

Abstract

Claims47