In pls.py, lines 148-151:
I am confused about why it is orig. for genes and z-score, but boot. for the p-value and adjusted p-value. As in genes.py, within the compute() function, _p_val and _p_corr seem to be ordered according to _index, which is the order of _z, rather than the original order. Although the two orders are similar, that is how the bootstrap process works.
In other words, in the output file pls_component_*.tsv, the order of the genes is the SAME as their order from the initial PLS regression result (in genes.py, line 120). This is really confusing.
Additionally, in genes.py, within the gsea function, the orig.genes is again used. If this is the case, why do we need to conduct the bootstrap of genes and get boot.genes?
I would appreciate any clarification on this matter.
In
pls.py, lines 148-151:I am confused about why it is
orig.for genes and z-score, butboot.for the p-value and adjusted p-value. As ingenes.py, within thecompute()function,_p_valand_p_corrseem to be ordered according to_index, which is the order of_z, rather than the original order. Although the two orders are similar, that is how the bootstrap process works.In other words, in the output file
pls_component_*.tsv, the order of the genes is the SAME as their order from the initial PLS regression result (ingenes.py, line 120). This is really confusing.Additionally, in
genes.py, within thegseafunction, theorig.genesis again used. If this is the case, why do we need to conduct the bootstrap of genes and getboot.genes?I would appreciate any clarification on this matter.