### Work on text

parent fea2cf0e
 ... ... @@ -135,15 +135,16 @@ Good, filtering for `Time_h` between 9.5 and 10.5 h gives a desired result. intervals? 10. There is one more thing. An intricacy. We have fitted and plotted (i.e. think about it) the `x`, the concentration, in logarithmic scale, but the think about) the `x`, the concentration, in logarithmic scale, but the IC~50~ is in linear scale. Mostly, it does not matter much. You can see above, however, that the lower confidence interval is 5x lower than IC~95~ and the upper limit is less than 2x higher. One side is 5 away, the other and the upper limit is less than 2x higher. One side is 5x away, the other less than 2x. To fix that, one could estimate IC~50~ in log scale (substitute IC~50~ in the 4-parameter logistic regression with log(IC~50~)). One might have to take some time to think about it what that means. Luckily, `drm` makes all this easy. You fit the model exactly as you did before, but this time, set `fct` to `LL2.4()`. Finally, when calculating MIC, the confidence interval should be set to "fls" (`interval = "fls"`). calculating MIC, the confidence interval should be set to "fls" (`interval = "fls"`). Now the MIC (IC~95~) should be the same (33 µM), but the confidence interval are symmetric, about 2x lower and 2x higher. Now the MIC (IC~95~) should be the same you got with `LL.4` (33 µM), but the confidence intervals are symmetric, about 2x lower and 2x higher.
 ... ... @@ -142,18 +142,19 @@ result. confidence intervals? 4. There is one more thing. An intricacy. We have fitted and plotted (i.e. think about it) the `x`, the concentration, in logarithmic scale, but the IC50 is in linear scale. Mostly, it does not matter much. You can see above, however, that the lower confidence interval is 5x lower than IC95 and the upper limit is less than 2x higher. One side is 5 away, the other less than 2x. To fix that, one could estimate IC50 in log scale (substitute IC50 in the 4-parameter logistic regression with log(IC50)). One might have to take some time to think about it what that means. Luckily, `drm` makes all this easy. You fit the model exactly as you did before, but this time, set `fct` to `LL2.4()`. Finally, when calculating MIC, the confidence interval should be set to “fls” (`interval = "fls"`). Now the MIC (IC95) should be the same (33 µM), but the confidence interval are symmetric, about 2x lower and 2x higher. (i.e. think about) the `x`, the concentration, in logarithmic scale, but the IC50 is in linear scale. Mostly, it does not matter much. You can see above, however, that the lower confidence interval is 5x lower than IC95 and the upper limit is less than 2x higher. One side is 5x away, the other less than 2x. To fix that, one could estimate IC50 in log scale (substitute IC50 in the 4-parameter logistic regression with log(IC50)). One might have to take some time to think about it what that means. Luckily, `drm` makes all this easy. You fit the model exactly as you did before, but this time, set `fct` to `LL2.4()`. Finally, when calculating MIC, the confidence interval should be set to “fls” (`interval = "fls"`). Now the MIC (IC95) should be the same you got with `LL.4` (33 µM), but the confidence intervals are symmetric, about 2x lower and 2x higher.
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment