r- mixtox tunefit函数以获取Beckon曲线拟合的初始估计值:参数为零

Question

我正在尝试使用MixTox软件包来执行Beckon Curve拟合。为了获得启动值，我使用tunefit（）函数。但是，我在下面提到的错误消息。我还尝试直接与我能想到的初始值拟合，但也会收到错误消息。有什么想法如何解决？

library(mixtox)
via_data$Concentration <- c(0.1974457, 0.3980034, 0.7921297, 1.5754719,
                            3.1546263, 6.2702489, 12.6250182, 25.2772970,
                            25.1103882, 49.9450449, 74.6800090)
via_data$Viability <- c(100.611247, 94.621027, 96.210269, 87.530562, 
                        80.562347,  62.224939,  39.119804, 
                        30.806846,  30.684597, 22.860636, 2.567237) 
via_tuneFit <- tuneFit(via_data$Concentration, via_data$Viability,
               eq="Beckon")
Error in if (nrow(param) > highBar) param <- param[sample(nrow(param),  : 
  argument is of length zero
via_curvFit <- curveFit(x=via_data$Concentration, rspn=via_data$Viability,
               eq="Beckon", rtype = "continuous", 
               param = c(77, 26, 1, 1, -6))
Processing continuous quantal dose-response data
Error in solve.default(t(jac) %*% jac) : 
  system is computationally singular: reciprocal condition number = 2.79026e-31

任何帮助将不胜感激。

谢谢Krina

Answer 1

这有效。

via_curvFit <- curveFit(x=via_data$Concentration, rspn=via_data$Viability,
               eq="Beckon", rtype = "continuous", param = c(1.32, 1.26, -0.94, 1.7, 0.68))

Answer 2

看来beckon方程不受Tunefit的支持。

这是TuneFit支持的：

models for curve fitting: "Hill", "Hill_two", "Hill_three", "Hill_four", "Weibull", "Weibull_three", "Weibull_four", "Logit", "Logit_three", "Logit_four", "BCW", "BCL", "GL", "Brain_Consens", "BCV", "Biphasic", "Hill_five".