Science means “to know” , which would extend to knowing about physical nature with us humans as observers and physical nature the object of observation. The following have been proposed for science to be better. They are as follows: Participatory observers replacing observers in the scientific method; determining lacunae in the present day methods used in scientific investigations; considering that present day scientists are in a Plato’s Cave ; circumventing mathematical equations in the scientific method via probabilities and cellular automaton; using technologically advanced aliens as an insightful lens, identifying the human blind spot; and determining what to sustain, eliminate and develop in future scientific human societies.
How could Science be Different
You make some proposals for science being better. In my opinion, science is like an algorithm, with logically constructing a theory and with testing it by experiments. What escapes this algorithm are the subjective opinions and assumptions of human beings. There will always be the problem of how to interpret experimental data, due to the encodings of different meanings within different subjective minds. The meaning attached to some data may be transformable algorithmically, but this does in no way result in some truth about the empirical world. In this sense you are correct about the participatory character of human beings and their facilitated machines.
However, the latter must be calibrated according to the assumptions of human beings to at all deliver an answer that could be consistent with what we already observed in nature. So there is a circulus vitiosus in play, independent of whether conscious observers are merely data processing devices or not: data can be biased by the biases of the agents that interpret them, be it AI machines or humans. And if conscious observers are not equal to data processing, then fundamental questions cannot be answered by merely data processing (like cellular automata), since this would be only a subset of what makes conscious observers be what they are.
This does not deny the usefullness of computers. But i did not understand what you want to say with your paragraph about signal transduction in human cells. So i ask: do you think that humans are just data processing devices (with some consciousness attached to that process)? Making science better crucially hinges at whether we believe that we are superior to logically interpret the outputs of some data processing machine or not. In the latter case there is the danger of no more scrutinizing whether or not our interpretations of the outputs are at all logically consistent.