Biomarkers for diagnosing non-alcoholic steatohepatitis (NASH) or hepatocellular carcinoma (HCC)
Embodiments described herein generally relate to technologies for analyzing peptide structures for diagnosing and/or treating a disease state advancing through a disease progression. A non-limiting example of a method relating to the technologies described in the subject application may include receiving peptide structure data corresponding to the biological sample obtained from the subject, identifying a peptide structure profile, and diagnosing a disease state within a disease progression. The example may further include generating a diagnosis output relating to the disease state. In at least some cases, the peptide structure profile may include glycosylated peptides, aglycosylated peptides, or both.
1. A method of classifying a biological sample with respect to a plurality of states associated with fatty liver disease (FLD) progression, the method comprising:
receiving peptide structure data corresponding to a set of glycoproteins in the biological sample obtained from a subject;
inputting quantification data identified from the peptide structure data for a set of peptide structures into a machine learning model, wherein the set of peptide structures includes at least one peptide structure identified from a plurality of peptide structures in Table 5, wherein the machine learning model is a regression model;
analyzing the quantification data using the machine learning model to generate a disease indicator; and generating a diagnosis output based on the disease indicator that classifies the biological sample as evidencing a corresponding state of the plurality of states associated with the FLD progression.
2. The method of claim 1 , wherein the disease indicator comprises a score and wherein generating the diagnosis output comprises:
determining that the score falls within a selected range associated with the corresponding state of the plurality of states; and
determining that the biological sample evidences the corresponding state in response to a determination that the score falls within the selected range associated with corresponding state.
3. The method of claim 1 , wherein the plurality of states includes a non-alcoholic steatohepatitis (NASH) state and a hepatocellular carcinoma (HCC) state.
4. The method of claim 1 , wherein the plurality of states includes a non-NASH/HCC state that comprises at least one of a healthy state, a liver disease-free state, or a benign hepatic mass state.
5. The method of claim 1 , wherein the at least one peptide structure comprises a glycopeptide structure defined by a peptide sequence and a glycan structure linked to the peptide sequence at a linking site of the peptide sequence, as identified in Table 5, with the peptide sequence being one of SEQ ID NOS: 23, 24, 25, 29, 30, 31, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, or 52 as defined in Table 5.
6. The method of claim 1 , wherein the quantification data for a peptide structure of the set of peptide structures comprises at least one of an abundance, a relative abundance, a normalized abundance, a relative quantity, an adjusted quantity, a normalized quantity, a relative concentration, an adjusted concentration, or a normalized concentration.
7. The method of claim 1 , wherein the peptide structure data is generated using multiple reaction monitoring mass spectrometry (MRM-MS).
8. A method of detecting a presence of one of a plurality of states associated with fatty liver disease (FLD) progression in a biological sample, the method comprising:
receiving peptide structure data corresponding to a set of glycoproteins in the biological sample obtained from a subject;
analyzing the peptide structure data using a supervised machine learning model to generate a disease indicator based on at least 3 peptide structures selected from a group of peptide structures identified in Table 5, wherein the machine learning model is a regression model; and
detecting the presence of a corresponding state of the plurality of states associated with the FLD progression in response to a determination that the disease indicator falls within a selected range associated with the corresponding state.
9. The method of claim 8 , wherein a peptide structure of the at least 3 peptide structures comprises a glycopeptide structure defined by a peptide sequence and a glycan structure linked to the peptide sequence at a linking site of the peptide sequence, as identified in Table 5, with the peptide sequence being one of SEQ ID NOS: 23, 24, 25, 29, 30, 31, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 52, or 52 as defined in Table 5.
10. The method of claim 8 , wherein the supervised machine learning model comprises a logistic regression model.
11. The method of claim 8 , wherein the supervised machine learning model comprises a penalized multivariable logistic regression model.
12. The method of claim 8 , wherein the peptide structure data comprises quantification data, the quantification data for a peptide structure of the group of peptide structures comprising at least one of an abundance, a relative abundance, a normalized abundance, a relative quantity, an adjusted quantity, a normalized quantity, a relative concentration, an adjusted concentration, or a normalized concentration.
13. The method of claim 8 , wherein the disease indicator is a probability score.
14. The method of claim 8 , further comprising:
generating a report that includes a diagnosis based on the corresponding state detected for the subject.
15. The method of claim 8 , wherein the plurality of states includes at least two selected from a group consisting of a non-alcoholic steatohepatitis (NASH) state, a hepatocellular carcinoma (HCC) state, and a non-NASH/HCC state.
16. The method of claim 1 , wherein the regression model comprises a LASSO model, or a LASSO regularization model.
17. The method of claim 16 , wherein the regression model is trained to compute a disease indicator or identify weight coefficients for peptide structures of a set of peptide structures.
18. The method of claim 8 , wherein the regression model is a LASSO model, or a LASSO regularization model.
19. The method of claim 18 , wherein the regression model is trained to compute a disease indicator or identify weight coefficients for peptide structures of a set of peptide structures.