Hello! I am working with longitudinal peptidomics data and would appreciate some advice on the most appropriate statistical approach.

I have previously worked with standard differential expression analysis, but not in a setting with repeated measurements across multiple timepoints, so I am unsure about the best way to handle this.

My dataset contains proteomic measurements from patients belonging to two clinical groups (Disease vs not), measured at multiple timepoints (for example H0, H24, H48). My main goal is to identify proteins that are differentially expressed between the two conditions.

My current idea is to fit, for each protein, a linear mixed-effects model of the form:

protein ~ Disease * timepoint + Age + Sex + Diabetes + (1 | Patient)

and then use contrasts to compare Disease vs Not within each timepoint, for example:

H0: Disease vs Non-Disease
H4:  Disease vs Non-Disease
H24:  Disease vs Non-Disease

My questions are:

1. Does this framework make sense for identifying differentially expressed proteins between groups at each timepoint?
2. Is it statistically appropriate to extract timepoint-specific contrasts from this mixed model and then apply multiple-testing correction across proteins within each timepoint?
3. Would there be a more standard or statistically preferable approach for this kind of longitudinal differential expression analysis in peptidomics/proteomics?

Any advice on best practices, or recommended packages/workflows would be very helpful. Thank you!