I am having trouble identifying the most appropriate statistical test for an experiment I am designing. In this experiment I am looking at multiple markers that indicate a change in structure using area measurements and histology. I have N = 6 mice and n = approx. 5-15 ROIs and 4 treatment groups (Control, Disease, Disease + New Drug and Disease + Current standard of care). A limitation I cannot control for is spatial hierarchy so I must assume the size of the object is proportional to the baseline expression of each marker. There is a fair amount of variation within each animal as well as within each treatment group.
The first problem I encounter is avoiding averaging out any of my data to focus on the variation between biological replicates as this may cause me to overlook important variation between ROIs in my analysis, particularly important I feel due to unknown spatial hierarchy.
The second problem I have encountered is that I have measured a number of markers where an increase from control suggests pathological restructuring. In isolation the changes in each marker are not dramatic as this is a short-term experiment. Each marker indicates restructuring by potentially different mechanisms. Furthermore, the markers are not independent from one another as they all contribute to the disease. This suggests to me that I should consider all of the changes in a single analysis, where more restructuring = worsened disease progression, opposed to multiple ANOVAs. Would a repeated measures ANOVA or Friedman test be applicable here?
I am unsure which test would best accommodate for this. Any suggestions would be greatly appreciated.
Example of my data:
Treatment Group Animal Structure size Marker 1 Marker 2 Marker 3...
Control 1 1000 20 40 90
2 750
Disease 3 800
4 1050
Disease + Drug 5 750
6 830
Disease + Old Drug 7 850
8 980 etc...
IMO We need a better description of the data and the research question. I would then look at: https://www.amazon.com/gp/product/1119593409/ref=s9_acsd_simh_bw_c_x_1_w?pf_rd_m=ATVPDKIKX0DER&pf_rd_s=merchandised-search-12&pf_rd_r=Q57KKQB003GJWV94H8FJ&pf_rd_t=101&pf_rd_p=889f20ae-7a86-4a16-ac38-93a1a337bd26&pf_rd_i=283155 and
perhaps also see the University Statistical Consultant. Best wishes, D. Booth
Have a look at this Q&A and especially the 2nd reply by David: https://www.researchgate.net/post/Whats_the_difference_between_Factorial_Experiment_Factorial_Design_and_Design_of_Experiments
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology
- Pharmaceuticals
- Drugs