A frame is a frame. This is going to raise a few eyebrows and probably start an argument or two in the thread but if blending a few frames in the original image sequence footage is acceptable then all you have to do is change the frame rate of the master comp you are using for the final render. There is absolutely no reason to change the frame rate of every comp or reinterpret the imported image sequences. If you just have to deliver all of the original frames in the image sequence to the client then all you have to do is render the master comp as an image sequence and send it to the client. Please let me explain.
If the project does was not supposed to be exactly 30 seconds for example and there is no sync sound then just render an image sequence and you are good to go. Just deliver an image sequence. If you need to deliver a 23.976 fps video then import the image sequence into Premiere Pro, After Effects or the Adobe Media Encoder and interpret the footage at 23.976 fps to render a video. Your animation will just take a bit longer to playback but you will not loose or blend any frames.
If you have sync sound or the animation is in sync with the imported video (image sequence) then using Mylenium's script will only help you change the frame rate of each comp. The script will not keep animation synec to the timing of the action in the imported footage if you re-interpret the footage and change the frame rate. Here is why. If you put a 30 second audio track in a composition that is 10 fps the audio will be 30 seconds. Put that same audio track in a composition that is 60fps and it will still be 30 seconds, so changing the frame rate of a composition will not change the speed of the sound. Changing the frame rate of a composition does not change the playback speed of footage placed in that composition even if it is an image sequence. Changing the interpreted frame rate of footage always changs the speed of the playback.
Now let's talk about animation. If you have your composition set to 30fps and you set a keyframe at 10 seconds and another keyframe at 20 seconds, then you go back and change the composition frame rate to 23.976 fps or 15 fps or 60fps, the keyframes will still be at 10 seconds and 20 seconds because keyframes are based on time, not frames. Changing the frame rate of the master comp will not change the timing of anything in any nested comp
So if your timing to the imported video is critical and you have sync sound then re-interpreting your image sequences is going to change the playback speed of that footage only. You may be better off just changing the frame rate of the composition and letting some frame blending happen when you render. That's what has been happening in NTSC countries since TV began when they put movies on TV 24 fps footage has to be interlaced and frame blended using 3:2 pulldown to keep the timing accurate and avoid the slowdown and the result is not unacceptable.
This is kind of a hard thing to wrap your head around and I hope I've explained it correctly. If you have sync sound and you absolutely must not blend any frames changing the interpretation of your footage (image sequence) will require that you adjust the timing of everything else in your composition and this could be a big can of worms. If you just need to deliver a modified copy of the original image sequence then you don't have to do anything but render a new image sequence. I will say it again. If blending a few frames in the original image sequence footage is acceptable then all you have to do is change the frame rate of the master comp you are using for the final render. There is no need to change the frame rate of every nested comp in the project. I don't do this all the time but I have mixed frame rate nested comps and footage in projects all the time and most of the time, blending frames is no problem at all.
If you are still confused please clarify the exact requirements of your project and let us know.