[FEATURE] Analysis based on the brightness of the state rather than the state label

[coltonbh]

I would like to have a tool that can make the analysis based on the brightness of the state, rather than the state label (AKA I want to be able to check the lifetime on the bright state, instead of the S2 state: I have a case where S1 and S2 are so close together that they continually swap in order along the simulations, which makes it hard to distinguish with the standard tools) @elipieri

[coltonbh]

@elipieri If you have any more details for this feature please put them here! Then we can track what it is you are hoping for, have conversations about any details we may need, and tie this all to code we will write :)

[elipieri]

Ok, more details: I have a system (thymine) where, at the hhTDA level of theory, the first two excited states are nearly energetically degenerate. One is bright, with a pi-pi* character, and one is dark, with a pi-sigma* character. Along the trajectory, the order of the states changes often. These characteristics make the trajectory spawn like crazy, but when you analyze the results with FAT (I never used aimsprop, but from what I understand it is similar), you can only plot the population by states labeled as S1 and S2. I, instead, need to plot the population of states identified as bright and dark. In other words, I want to make a population graphic that shows the lifetime of the bright state and the dark state, not S1 and S2, as they swap. If I am not explaining myself, let me know and I can show you my data. Elisa

…

On Fri, Feb 12, 2021 at 6:57 PM Colton Hicks ***@***.***> wrote: @elipieri <https://github.com/elipieri> If you have any more details for this feature please put them here! Then we can track what it is you are hoping for, have conversations about any details we may need, and tie this all to code we will write :) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ANWPUI4LBBFSZGU7NUVCOP3S6XTCVANCNFSM4XRV2CBQ> .

[walkerar]

We should be able to compute the oscillator strength from the transition dipole moment (TDip file) of each state at each timepoint, and extract the data for the largest value every step. The easiest hack for this would be to simply put this data into a combined new vector with a new state ID that is continuous, and then putting that back into the rest of aimsprop. So first:

1. Does aimsprop already read in the transition dipole moments with state labels
2. What is the correct math to convert TDip to oscillator strength and identify appropriate state at each time
3. Must write a small module to take the data at each time for the bright state and reformat it into a new vector that can be fed to aimsprop as normal