In our present study, we experimentally induced and collected 45 samples showing a wide range of clinical signs of digestive disorders, including the complete healthy, slight, mild, and severe individuals. However, it is actually hard to definitely classify them into different groups. The trouble appeared when I tried to detect the differential expression gene based on comprehensive RNA-seq for these samples; mostly because there was no clear definition of categorical groups/conditions as wildly adopted in classical RNA-seq analyses. Even when these individuals were carefully grouped, it is also reasonable that the total variance within group would be very high because these members are not strictly comparable to the general biological replicates. Therefore, my inquiry is that whether there is an appropriate statistical method or tool to detect the gene expression significantly associated with the continuous biological changes rather than the clear categorical groups, especially for RNA-seq.
