Techniques for bias correction in sequence data

Inventors

Nuzhdina, Ekaterina • Bagaev, Alexander • Chelushkin, Maksim • Lozinsky, Yaroslav • Miheecheva, Natalia • Zaitsev, Aleksandr

Assignees

BostonGene Corp

Abstract

Described herein are various methods of collecting and processing of tumor and/or healthy tissue samples to extract nucleic acid and perform nucleic acid sequencing. Also described herein are various methods of processing nucleic acid sequencing data to remove bias from the nucleic acid sequencing data. Also described herein are various methods of evaluating the quality of nucleic acid sequence information. The identity and/or integrity of nucleic acid sequence data is evaluated prior to using the sequence information for subsequent analysis (for example for diagnostic, prognostic, or clinical purposes). The methods enable a subject, doctor, or user to characterize or classify various types of cancer precisely, and thereby determine a therapy or combination of therapies that may be effective to treat a cancer in a subject based on the precise characterization.

Core Innovation

The invention provides a molecular functional (MF) profile framework in which cancer malignancy and cancer microenvironment are represented by gene groups whose expression levels are used to determine molecular functional profiles for a subject. The validated nucleic acid data are obtained from a previously obtained biological sample of a subject having cancer by performing MHC allele analysis, tumor type classification, and RNA-seq type classification, and identifying the nucleic acid data as validated nucleic acid data when the obtained information matches the asserted information.

The MF profile is associated with an MF profile cluster among multiple clusters generated by clustering MF profiles determined from RNA expression data for a plurality of subjects. The MF profile clusters are associated with combinations of inflamed/non-inflamed and vascularized/non-vascularized biological samples, including immune desert, and also with inflamed and fibroblast-enriched biological samples or inflamed and non-fibroblast-enriched biological samples.

The validated nucleic acid data are used as input for determining the MF profile for the subject by determining gene group expression levels for a gene group associated with cancer malignancy and a different gene group associated with cancer microenvironment, where the gene groups are defined by specified genes. The method includes clustering the MF profile of the subject with the plurality of MF profile clusters and determining that the subject has an inflamed/non-vascularized MF profile.

After determining the inflamed/non-vascularized MF profile, the method administers immune checkpoint blockade therapy to the subject having the inflamed/non-vascularized MF profile. This integrates validated nucleic-acid sequence data, gene-group expression level determination, and MF profile cluster assignment tied to inflammation and vascularization states.

Claims Coverage

The consolidated content provides one independent claim with 4 inventive feature groupings. It combines validation of nucleic-acid sequence data, MF profile determination from specified gene groups, MF profile cluster assignment by inflammation and vascularization states, and administration of immune checkpoint blockade therapy based on an assigned MF profile cluster.

The independent claim ties together validated nucleic-acid sequence data, MF profile determination from specified gene groups, MF profile cluster assignment by inflammation and vascularization states, and administration of immune checkpoint blockade therapy for an inflamed/non-vascularized MF profile.

Stated Advantages

Documented Applications