While investigating some peak picking automation, i found that the value of MI, which is in relative units, displays an odd relationship to the value of ymax_p. I had assumed that since MI can only be set in relative units, that these units would be a proportion of the magnitude of the data, such that ymax_p would be a relative value 1.0, or, even the range of ymin_p - ymax_p would be relative 0 and 1, however this is not the case.

I was hoping to be able to convert between the two values, but there seems to be another factor in the relationship between the relative and absolute values, at least with respect to MI being a cutoff for peak picking.

I attempted to find this [assumed] scalar value, thinking that perhaps it had a relationship to the s_dev of the data, but have not been able to establish a direct relationship between the two.

• Does anyone know what the relationship between the [abs] and [rel] are?

• As a side question, how is s_dev calculated, and what is it the actual standard deviation of?

Cheers and thanks,

Joseph