Methods and compositions of molecular profiling for disease diagnostics

A method of evaluating a sample of thyroid tissue from a patient comprising the steps of:

• a. obtaining said sample of thyroid tissue from said patient;
• b. assaying an expression level for two or more gene expression products in said sample of thyroid tissue to generate test data
• c. correlating said test data with gene expression data obtained from at least 200 training samples of thyroid tissue;
• d. based on said correlating, (i) identifying said sample of thyroid tissue as malignant, (ii) identifying said sample of thyroid tissue as benign, (iii) identifying said sample of thyroid tissue as non-cancerous, (iv) identifying said sample of thyroid tissue as non-malignant, or (v) identifying said sample of thyroid tissue as normal, wherein said correlating uses gene expression data of at least two of the following genes or corresponding Transcript Cluster ID Nos:
• ARSG, FREM2, GIMAP2, PIGN, and PLA2G16;
• and
• e. treating said patient on the basis of step (d),
• wherein one or more of said at least 200 training samples of thyroid tissues have a pathology selected from the group consisting of:
• metastatic melanoma, metastatic renal carcinoma, metastatic breast carcinoma, and metastatic B cell lymphoma.

A method of evaluating a sample of thyroid tissue from a patient comprising the steps of:

• a. obtaining said sample of thyroid tissue from said patient;
• b. assaying an expression level for two or more gene expression products in said sample of thyroid tissue;
• c. applying an algorithm to said expression level of step (b), wherein said algorithm correlates the expression level of step (b) with expression data from at least three samples of thyroid tissue, wherein each of said at least three samples of thyroid tissue is obtained from a different tissue type and wherein said different tissue type is selected from the following tissue types:
• follicular carcinoma, lymphocytic thyroiditis, follicular variant papillary thyroid carcinoma, papillary thyroid carcinoma, nodular hyperplasia, medullary thyroid carcinoma, Hurthle cell carcinoma, Hurthle cell adenoma, anaplastic thyroid carcinoma, metastatic melanoma, metastatic renal carcinoma, metastatic breast carcinoma, parathyroid, and metastatic B cell lymphoma;
• d. using said algorithm, (i) identifying said sample of thyroid tissue as malignant, (ii) identifying said sample of thyroid tissue as benign, (iii) identifying said sample of thyroid tissue as non-cancerous, (iv) identifying said sample of thyroid tissue as non-malignant, or (v) identifying said sample of thyroid tissue as normal, wherein said expression data include data of at least two of the following genes or corresponding Transcript Cluster ID Nos:
• ARSG, FREM2, GIMAP2, PIGN, and PLA2G16;
• and
• e. treating said patient on the basis of step (d),
• wherein said algorithm is a trained algorithm, trained by said expression data from at least three samples of thyroid tissue, wherein each of said samples of thyroid tissue is obtained from a different tissue type and wherein said different tissue type is selected from the following tissue types:
• follicular carcinoma, lymphocytic thyroiditis, follicular variant papillary thyroid carcinoma, papillary thyroid carcinoma, nodular hyperplasia, medullary thyroid carcinoma, Hurthle cell carcinoma, Hurthle cell adenoma, anaplastic thyroid carcinoma, metastatic melanoma, metastatic renal carcinoma, metastatic breast carcinoma, parathyroid, or metastatic B cell lymphoma,
• and further wherein said training samples of thyroid tissue comprise three or more samples of the following types:
• metastatic melanoma, metastatic renal carcinoma, metastatic breast carcinoma, metastatic B cell lymphoma, or parathyroid.

A method of evaluating a sample of thyroid tissue from a patient comprising the steps of:

• a. obtaining said sample of thyroid tissue from said patient;
• b. assaying an expression level for two or more gene expression products in said sample of thyroid tissue to generate test data;
• c. applying an algorithm to said expression level of step (b), wherein said algorithm correlates the expression level of step (b) with expression data obtained from a plurality of samples, wherein said plurality of samples comprises a sample obtained by surgical biopsy of thyroid tissue and a sample obtained by fine needle aspiration of thyroid tissue;
• d. based on said correlating, (i) identifying said sample of thyroid tissue as malignant, (ii) identifying said sample of thyroid tissue as benign, (iii) identifying said sample of thyroid tissue as non-cancerous, (iv) identifying said sample of thyroid tissue as non-malignant, or (v) identifying said sample of thyroid tissue as normal, wherein said correlating uses gene expression data of at least two of the following genes or corresponding Transcript Cluster ID Nos:
• ARSG, FREM2, GIMAP2, PIGN, and PLA2G16;
• and
• e. treating said patient on the basis of step (d), wherein said algorithm is a trained algorithm trained by said expression data obtained from said plurality of samples, and further wherein one or more of said plurality of samples has a pathology selected from the group consisting of:
• metastatic melanoma, metastatic renal carcinoma, metastatic breast carcinoma, and metastatic B cell lymphoma.

A method of detecting a thyroid condition in a patient comprising the steps of:

• a. obtaining a sample of thyroid tissue from said patient;
• b. assaying a level of expression of two or more gene expression products in said sample of thyroid tissue, wherein the two or more gene assayed include at least two of the following genes or corresponding Transcript Cluster ID Nos:
• ARSG, FREM2, GIMAP2, PIGN, and PLA2G 16;
• c. classifying said sample of thyroid tissue as benign by applying an algorithm to data from step b, wherein said algorithm correlates said data from step b with expression data obtained from a plurality of samples, wherein said plurality of samples comprises a sample with a pathology that is a metastatic cancer from a non-thyroid organ, wherein said expression data include data of at least two of the following genes or corresponding Transcript Cluster ID Nos:
• ARSG, FREM2, GIMAP2, PIGN, and PLA2G16;
• and
• d. treating said patient on the basis of step (c).

A method of evaluating a sample of thyroid tissue from a patient comprising the steps of:

• a. obtaining said sample of thyroid tissue from said patient;
• b. assaying an expression level for two or more gene expression products in said sample of thyroid tissue;
• c. applying an algorithm to said expression level of step (b), wherein said algorithm correlates the expression level of step (b) with expression data from at least three samples of thyroid tissue, wherein each of said at least three samples of thyroid tissue is obtained from a different tissue type and wherein said different tissue type is selected from the following tissue types:
• follicular carcinoma, lymphocytic thyroiditis, follicular variant papillary thyroid carcinoma, papillary thyroid carcinoma, nodular hyperplasia, medullary thyroid carcinoma, Hurthle cell carcinoma, Hurthle cell adenoma, anaplastic thyroid carcinoma, metastatic melanoma, metastatic renal carcinoma, metastatic breast carcinoma, parathyroid, and metastatic B cell lymphoma;
• d. using said algorithm, (i) identifying said sample of thyroid tissue as malignant, (ii) identifying said sample of thyroid tissue as benign, (iii) identifying said sample of thyroid tissue as non-cancerous, (iv) identifying said sample of thyroid tissue as non-malignant, or (v) identifying said sample of thyroid tissue as normal, wherein said expression data include data of at least two of the following genes or corresponding Transcript Cluster ID Nos:
• ARSG, FREM2, GIMAP2, PIGN, and PLA2G16;
• and
• e. treating said patient on the basis of step (d),
• wherein the three samples of thyroid tissue are the following tissue types:
• follicular carcinoma, papillary thyroid carcinoma, and metastatic B cell lymphoma.