A steep slope may be caused by the IC50 being below the concentration of the receptor or enzyme to which it binds.
If the concentration range were much higher (micromolar instead of nanomolar), I would suggest that insolubility of the inhibitor could be responsible for artifacts, but this is unlikely at 10 nM.
The hook effect may only be apparent, due to experimental variation (noise). Repeating the measurement should clarify whether the phenomenon is reproducible.
Hi Zoe Meng Two questions: How can you get more than 100% inhibition in an assay? And there are no error bars. Why is the X-axis concentration in log? The slope will not be as steep if plotted conventionally.