See https://numpy.org/neps/nep-0051-scalar-representation.html - for example this:

>>> np.float32(3.0)
3.0

becomes:

>>> np.float32(3.0)
np.float32(3.0)

As per the NEP, this is expected to require documentation updates in libraries using numpy - like Biopython. Out of practicality we will probably do this alongside setting a minimum dependency of numpy 2.0.0