Should we always worry about getting correct and accurate results? Or identify all possible hits out there?
I believe this will never be possible, because of problems addressed here and also from inherent flaws in the force fields and scoring functions. Instead, if we treat such methods to find hits not easily recognisable to human insight ( of course this is subjective based on who is working and with what level of experience) and use this as a model to guide experiments, we would do much better. As medicinal chemistry program proceeds, going from hit to lead to clinical candidate, we learn more from experiments than theoretical work.
In the long run such results progressively fill the missing links for a particular target or program.
Should we always worry about getting correct and accurate results? Or identify all possible hits out there?
I believe this will never be possible, because of problems addressed here and also from inherent flaws in the force fields and scoring functions. Instead, if we treat such methods to find hits not easily recognisable to human insight ( of course this is subjective based on who is working and with what level of experience) and use this as a model to guide experiments, we would do much better. As medicinal chemistry program proceeds, going from hit to lead to clinical candidate, we learn more from experiments than theoretical work.
In the long run such results progressively fill the missing links for a particular target or program.