path: root/vignettes/web_only/dimethenamid_2018.rmd

---
title: Example evaluations of the dimethenamid data from 2018
author: Johannes Ranke
date: Last change 27 September 2021, built on `r format(Sys.Date(), format = "%d %b %Y")`
output:
  html_document:
    toc: true
    toc_float: true
    code_folding: hide
    fig_retina: null
bibliography: ../references.bib
vignette: >
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

[Wissenschaftlicher Berater, Kronacher Str. 12, 79639 Grenzach-Wyhlen, Germany](http://www.jrwb.de)

```{r, include = FALSE}
require(knitr)
require(mkin)
require(nlme)
options(digits = 5)
opts_chunk$set(
  comment = "",
  tidy = FALSE,
  cache = TRUE
)
```

# Introduction

During the preparation of the journal article on nonlinear mixed-effects models
in degradation kinetics [@ranke2021] and the analysis of the dimethenamid
degradation data analysed therein, a need for a more detailed analysis using
not only nlme and saemix, but also nlmixr for fitting the mixed-effects models
was identified, as many model variants do not converge when fitted with nlme,
and not all relevant error models can be fitted with saemix.

This vignette is an attempt to satisfy this need.

# Data

Residue data forming the basis for the endpoints derived in the conclusion on
the peer review of the pesticide risk assessment of dimethenamid-P published by
the European Food Safety Authority (EFSA) in 2018 [@efsa_2018_dimethenamid]
were transcribed from the risk assessment report [@dimethenamid_rar_2018_b8]
which can be downloaded from the Open EFSA repository
[https://open.efsa.europa.eu/study-inventory/EFSA-Q-2014-00716](https://open.efsa.europa.eu).

The data are [available in the mkin
package](https://pkgdown.jrwb.de/mkin/reference/dimethenamid_2018.html). The
following code (hidden by default, please use the button to the right to show
it) treats the data available for the racemic mixture dimethenamid (DMTA) and
its enantiomer dimethenamid-P (DMTAP) in the same way, as no difference between
their degradation behaviour was identified in the EU risk assessment. The
observation times of each dataset are multiplied with the corresponding
normalisation factor also available in the dataset, in order to make it
possible to describe all datasets with a single set of parameters.

Also, datasets observed in the same soil are merged, resulting in dimethenamid
(DMTA) data from six soils.

```{r dimethenamid_data}
library(mkin, quietly = TRUE)
dmta_ds <- lapply(1:7, function(i) {
  ds_i <- dimethenamid_2018$ds[[i]]$data
  ds_i[ds_i$name == "DMTAP", "name"] <-  "DMTA"
  ds_i$time <- ds_i$time * dimethenamid_2018$f_time_norm[i]
  ds_i
})
names(dmta_ds) <- sapply(dimethenamid_2018$ds, function(ds) ds$title)
dmta_ds[["Elliot"]] <- rbind(dmta_ds[["Elliot 1"]], dmta_ds[["Elliot 2"]])
dmta_ds[["Elliot 1"]] <- NULL
dmta_ds[["Elliot 2"]] <- NULL
```

# Parent degradation

We evaluate the observed degradation of the parent compound using simple
exponential decline (SFO) and biexponential decline (DFOP), using constant
variance (const) and a two-component variance (tc) as error models.

## Separate evaluations

As a first step, to get a visual impression of the fit of the different models,
we do separate evaluations for each soil using the mmkin function from the
mkin package:

```{r f_parent_mkin}
f_parent_mkin_const <- mmkin(c("SFO", "DFOP"), dmta_ds,
  error_model = "const", quiet = TRUE)
f_parent_mkin_tc <- mmkin(c("SFO", "DFOP"), dmta_ds,
  error_model = "tc", quiet = TRUE)
```

The plot of the individual SFO fits shown below suggests that at least in some
datasets the degradation slows down towards later time points, and that the
scatter of the residuals error is smaller for smaller values (panel to the
right):

```{r f_parent_mkin_sfo_const}
plot(mixed(f_parent_mkin_const["SFO", ]))
```

Using biexponential decline (DFOP) results in a slightly more random
scatter of the residuals:

```{r f_parent_mkin_dfop_const}
plot(mixed(f_parent_mkin_const["DFOP", ]))
```

The population curve (bold line) in the above plot results from taking the mean
of the individual transformed parameters, i.e. of log k1 and log k2, as well as
of the logit of the g parameter of the DFOP model). Here, this procedure
does not result in parameters that represent the degradation well, because in some
datasets the fitted value for k2 is extremely close to zero, leading to a log
k2 value that dominates the average. This is alleviated if only rate constants
that pass the t-test for significant difference from zero (on the untransformed
scale) are considered in the averaging:

```{r f_parent_mkin_dfop_const_test}
plot(mixed(f_parent_mkin_const["DFOP", ]), test_log_parms = TRUE)
```

While this is visually much more satisfactory, such an average procedure could
introduce a bias, as not all results from the individual fits enter the
population curve with the same weight. This is where nonlinear mixed-effects
models can help out by treating all datasets with equally by fitting a
parameter distribution model together with the degradation model and the error
model (see below).

The remaining trend of the residuals to be higher for higher predicted residues
is reduced by using the two-component error model:

```{r f_parent_mkin_dfop_tc_test}
plot(mixed(f_parent_mkin_tc["DFOP", ]), test_log_parms = TRUE)
```

## Nonlinear mixed-effects models

Instead of taking a model selection decision for each of the individual fits, we fit
nonlinear mixed-effects models (using different fitting algorithms as implemented in
different packages) and do model selection using all available data at the same time.
In order to make sure that these decisions are not unduly influenced by the
type of algorithm used, by implementation details or by the use of wrong control
parameters, we compare the model selection results obtained with different R
packages, with different algorithms and checking control parameters.

### nlme

The nlme package was the first R extension providing facilities to fit nonlinear
mixed-effects models. We would like to do model selection from all four
combinations of degradation models and error models based on the AIC.
However, fitting the DFOP model with constant variance and using default
control parameters results in an error, signalling that the maximum number
of 50 iterations was reached, potentially indicating overparameterisation.
Nevertheless, the algorithm converges when the two-component error model is
used in combination with the DFOP model. This can be explained by the fact
that the smaller residues observed at later sampling times get more
weight when using the two-component error model which will counteract the
tendency of the algorithm to try parameter combinations unsuitable for
fitting these data.

```{r f_parent_nlme, warning = FALSE}
library(nlme)
f_parent_nlme_sfo_const <- nlme(f_parent_mkin_const["SFO", ])
# f_parent_nlme_dfop_const <- nlme(f_parent_mkin_const["DFOP", ])
f_parent_nlme_sfo_tc <- nlme(f_parent_mkin_tc["SFO", ])
f_parent_nlme_dfop_tc <- nlme(f_parent_mkin_tc["DFOP", ])
```

Note that a certain degree of overparameterisation is also indicated by a
warning obtained when fitting DFOP with the two-component error model ('false
convergence' in the 'LME step' in iteration 3). However, as this warning does
not occur in later iterations, and specifically not in the last of the
`r f_parent_nlme_dfop_tc$numIter` iterations, we can ignore this warning.

The model comparison function of the nlme package can directly be applied
to these fits showing a much lower AIC for the DFOP model fitted with the
two-component error model. Also, the likelihood ratio test indicates that this
difference is significant as the p-value is below 0.0001.

```{r AIC_parent_nlme}
anova(
  f_parent_nlme_sfo_const, f_parent_nlme_sfo_tc, f_parent_nlme_dfop_tc
)
```

In addition to these fits, attempts were also made to include correlations
between random effects by using the log Cholesky parameterisation of the matrix
specifying them. The code used for these attempts can be made visible below.

```{r f_parent_nlme_logchol, warning = FALSE, eval = FALSE}
f_parent_nlme_sfo_const_logchol <- nlme(f_parent_mkin_const["SFO", ],
  random = nlme::pdLogChol(list(DMTA_0 ~ 1, log_k_DMTA ~ 1)))
anova(f_parent_nlme_sfo_const, f_parent_nlme_sfo_const_logchol)
f_parent_nlme_sfo_tc_logchol <- nlme(f_parent_mkin_tc["SFO", ],
  random = nlme::pdLogChol(list(DMTA_0 ~ 1, log_k_DMTA ~ 1)))
anova(f_parent_nlme_sfo_tc, f_parent_nlme_sfo_tc_logchol)
f_parent_nlme_dfop_tc_logchol <- nlme(f_parent_mkin_const["DFOP", ],
  random = nlme::pdLogChol(list(DMTA_0 ~ 1, log_k1 ~ 1, log_k2 ~ 1, g_qlogis ~ 1)))
anova(f_parent_nlme_dfop_tc, f_parent_nlme_dfop_tc_logchol)
```

While the SFO variants converge fast, the additional parameters introduced
by this lead to convergence warnings for the DFOP model. The model comparison
clearly show that adding correlations between random effects does not improve
the fits.

The selected model (DFOP with two-component error) fitted to the data assuming
no correlations between random effects is shown below.

```{r plot_parent_nlme}
plot(f_parent_nlme_dfop_tc)
```

### saemix

The saemix package provided the first Open Source implementation of the
Stochastic Approximation to the Expectation Maximisation (SAEM) algorithm.
SAEM fits of degradation models can be conveniently performed using an
interface to the saemix package available in current development versions of
the mkin package.

The corresponding SAEM fits of the four combinations of degradation and error
models are fitted below. As there is no convergence criterion implemented in
the saemix package, the convergence plots need to be manually checked for every
fit. As we will compare the SAEM implementation of saemix to the results
obtained using the nlmixr package later, we define control settings that
work well for all the parent data fits shown in this vignette.

```{r saemix_control}
library(saemix)
saemix_control <- saemixControl(nbiter.saemix = c(800, 300), nb.chains = 15,
    print = FALSE, save = FALSE, save.graphs = FALSE, displayProgress = FALSE)
saemix_control_10k <- saemixControl(nbiter.saemix = c(10000, 1000), nb.chains = 15,
    print = FALSE, save = FALSE, save.graphs = FALSE, displayProgress = FALSE)
```

The convergence plot for the SFO model using constant variance is shown below.

```{r f_parent_saemix_sfo_const, results = 'hide', dependson = "saemix_control"}
f_parent_saemix_sfo_const <- mkin::saem(f_parent_mkin_const["SFO", ], quiet = TRUE,
  control = saemix_control, transformations = "saemix")
plot(f_parent_saemix_sfo_const$so, plot.type = "convergence")
```

Obviously the default number of iterations is sufficient to reach convergence.
This can also be said for the SFO fit using the two-component error model.

```{r f_parent_saemix_sfo_tc, results = 'hide', dependson = "saemix_control"}
f_parent_saemix_sfo_tc <- mkin::saem(f_parent_mkin_tc["SFO", ], quiet = TRUE,
  control = saemix_control, transformations = "saemix")
plot(f_parent_saemix_sfo_tc$so, plot.type = "convergence")
```

When fitting the DFOP model with constant variance (see below), parameter
convergence is not as unambiguous.

```{r f_parent_saemix_dfop_const, results = 'hide', dependson = "saemix_control"}
f_parent_saemix_dfop_const <- mkin::saem(f_parent_mkin_const["DFOP", ], quiet = TRUE,
  control = saemix_control, transformations = "saemix")
plot(f_parent_saemix_dfop_const$so, plot.type = "convergence")
```

This is improved when the DFOP model is fitted with the two-component error
model. Convergence of the variance of k2 is enhanced, it remains more or less
stable already after 200 iterations of the first phase.

```{r f_parent_saemix_dfop_tc, results = 'hide', dependson = "saemix_control"}
f_parent_saemix_dfop_tc <- mkin::saem(f_parent_mkin_tc["DFOP", ], quiet = TRUE,
  control = saemix_control, transformations = "saemix")
plot(f_parent_saemix_dfop_tc$so, plot.type = "convergence")
```

We also check if using many more iterations (10 000 for the first and 1000 for
the second phase) improve the result in a significant way. The AIC values
obtained are compared further below.

```{r f_parent_saemix_dfop_tc_10k, results = 'hide', dependson = "saemix_control"}
f_parent_saemix_dfop_tc_10k <- mkin::saem(f_parent_mkin_tc["DFOP", ], quiet = TRUE,
  control = saemix_control_10k, transformations = "saemix")
plot(f_parent_saemix_dfop_tc_10k$so, plot.type = "convergence")
```

An alternative way to fit DFOP in combination with the two-component error model
is to use the model formulation with transformed parameters as used per default
in mkin.

```{r f_parent_saemix_dfop_tc_mkin, results = 'hide', dependson = "saemix_control"}
f_parent_saemix_dfop_tc_mkin <- mkin::saem(f_parent_mkin_tc["DFOP", ], quiet = TRUE,
  control = saemix_control, transformations = "mkin")
plot(f_parent_saemix_dfop_tc_mkin$so, plot.type = "convergence")
```

As the convergence plots do not clearly indicate that the algorithm has converged, we
again use a much larger number of iterations, which leads to satisfactory
convergence (see below).

```{r f_parent_saemix_dfop_tc_mkin_10k, results = 'hide', dependson = "saemix_control"}
f_parent_saemix_dfop_tc_mkin_10k <- mkin::saem(f_parent_mkin_tc["DFOP", ], quiet = TRUE,
  control = saemix_control_10k, transformations = "mkin")
plot(f_parent_saemix_dfop_tc_mkin_10k$so, plot.type = "convergence")
```

The four combinations (SFO/const, SFO/tc, DFOP/const and DFOP/tc), including
the variations of the DFOP/tc combination can be compared using the model
comparison function of the saemix package:

```{r AIC_parent_saemix, cache = FALSE}
AIC_parent_saemix <- saemix::compare.saemix(
  f_parent_saemix_sfo_const$so,
  f_parent_saemix_sfo_tc$so,
  f_parent_saemix_dfop_const$so,
  f_parent_saemix_dfop_tc$so,
  f_parent_saemix_dfop_tc_10k$so,
  f_parent_saemix_dfop_tc_mkin$so,
  f_parent_saemix_dfop_tc_mkin_10k$so)
rownames(AIC_parent_saemix) <- c(
  "SFO const", "SFO tc", "DFOP const", "DFOP tc", "DFOP tc more iterations",
  "DFOP tc mkintrans", "DFOP tc mkintrans more iterations")
print(AIC_parent_saemix)
```

As in the case of nlme fits, the DFOP model fitted with two-component error
(number 4) gives the lowest AIC. Using a much larger number of iterations
does not improve the fit a lot. When the mkin transformations are used
instead of the saemix transformations, this large number of iterations leads
to a goodness of fit that is comparable to the result obtained with saemix
transformations.

In order to check the influence of the likelihood calculation algorithms
implemented in saemix, the likelihood from Gaussian quadrature is added
to the best fit, and the AIC values obtained from the three methods
are compared.

```{r AIC_parent_saemix_methods, cache = FALSE}
f_parent_saemix_dfop_tc$so <-
  saemix::llgq.saemix(f_parent_saemix_dfop_tc$so)
AIC_parent_saemix_methods <- c(
  is = AIC(f_parent_saemix_dfop_tc$so, method = "is"),
  gq = AIC(f_parent_saemix_dfop_tc$so, method = "gq"),
  lin = AIC(f_parent_saemix_dfop_tc$so, method = "lin")
)
print(AIC_parent_saemix_methods)
```

The AIC values based on importance sampling and Gaussian quadrature are very
similar. Using linearisation is known to be less accurate, but still gives a
similar value.

### nlmixr

In the last years, a lot of effort has been put into the nlmixr package which
is designed for pharmacokinetics, where nonlinear mixed-effects models are
routinely used, but which can also be used for related data like chemical
degradation data. A current development branch of the mkin package provides
an interface between mkin and nlmixr. Here, we check if we get equivalent
results when using a refined version of the First Order Conditional Estimation
(FOCE) algorithm used in nlme, namely the First Order Conditional Estimation
with Interaction (FOCEI), and the SAEM algorithm as implemented in nlmixr.

First, the focei algorithm is used for the four model combinations. A number of
warnings are produced with unclear significance.

```{r f_parent_nlmixr_focei, results = "hide", message = FALSE, warning = FALSE}
library(nlmixr)
f_parent_nlmixr_focei_sfo_const <- nlmixr(f_parent_mkin_const["SFO", ], est = "focei")
f_parent_nlmixr_focei_sfo_tc <- nlmixr(f_parent_mkin_tc["SFO", ], est = "focei")
f_parent_nlmixr_focei_dfop_const <- nlmixr(f_parent_mkin_const["DFOP", ], est = "focei")
f_parent_nlmixr_focei_dfop_tc<- nlmixr(f_parent_mkin_tc["DFOP", ], est = "focei")
```

```{r AIC_parent_nlmixr_focei, cache = FALSE}
aic_nlmixr_focei <- sapply(
  list(f_parent_nlmixr_focei_sfo_const$nm, f_parent_nlmixr_focei_sfo_tc$nm,
    f_parent_nlmixr_focei_dfop_const$nm, f_parent_nlmixr_focei_dfop_tc$nm),
  AIC)
```

The AIC values are very close to the ones obtained with nlme which are repeated below
for convenience.

```{r AIC_parent_nlme_rep, cache = FALSE}
aic_nlme <- sapply(
  list(f_parent_nlme_sfo_const, NA, f_parent_nlme_sfo_tc, f_parent_nlme_dfop_tc),
  function(x) if (is.na(x[1])) NA else AIC(x))
aic_nlme_nlmixr_focei <- data.frame(
  "Degradation model" = c("SFO", "SFO", "DFOP", "DFOP"),
  "Error model" = rep(c("constant variance", "two-component"), 2),
  "AIC (nlme)" = aic_nlme,
  "AIC (nlmixr with FOCEI)" = aic_nlmixr_focei,
  check.names = FALSE
)
```

Secondly, we use the SAEM estimation routine and check the convergence plots. The
control parameters also used for the saemix fits are defined beforehand.

```{r nlmixr_saem_control}
nlmixr_saem_control_800 <- saemControl(logLik = TRUE,
  nBurn = 800, nEm = 300, nmc = 15)
nlmixr_saem_control_1000 <- saemControl(logLik = TRUE,
  nBurn = 1000, nEm = 300, nmc = 15)
nlmixr_saem_control_10k <- saemControl(logLik = TRUE,
  nBurn = 10000, nEm = 1000, nmc = 15)
```

The we fit SFO with constant variance

```{r f_parent_nlmixr_saem_sfo_const, results = "hide", warning = FALSE, message = FALSE, dependson = "nlmixr_saem_control"}
f_parent_nlmixr_saem_sfo_const <- nlmixr(f_parent_mkin_const["SFO", ], est = "saem",
  control = nlmixr_saem_control_800)
traceplot(f_parent_nlmixr_saem_sfo_const$nm)
```

and SFO with two-component error.

```{r f_parent_nlmixr_saem_sfo_tc, results = "hide", warning = FALSE, message = FALSE, dependson = "nlmixr_saem_control"}
f_parent_nlmixr_saem_sfo_tc <- nlmixr(f_parent_mkin_tc["SFO", ], est = "saem",
  control = nlmixr_saem_control_800)
traceplot(f_parent_nlmixr_saem_sfo_tc$nm)
```

For DFOP with constant variance, the convergence plots show considerable
instability of the fit, which indicates overparameterisation which was already
observed earlier for this model combination.

```{r f_parent_nlmixr_saem_dfop_const, results = "hide", warning = FALSE, message = FALSE, dependson = "nlmixr_saem_control"}
f_parent_nlmixr_saem_dfop_const <- nlmixr(f_parent_mkin_const["DFOP", ], est = "saem",
  control = nlmixr_saem_control_800)
traceplot(f_parent_nlmixr_saem_dfop_const$nm)
```

For DFOP with two-component error, a less erratic convergence is seen.

```{r f_parent_nlmixr_saem_dfop_tc, results = "hide", warning = FALSE, message = FALSE, dependson = "nlmixr_saem_control"}
f_parent_nlmixr_saem_dfop_tc <- nlmixr(f_parent_mkin_tc["DFOP", ], est = "saem",
  control = nlmixr_saem_control_800)
traceplot(f_parent_nlmixr_saem_dfop_tc$nm)
```

To check if an increase in the number of iterations improves the fit, we repeat
the fit with 1000 iterations for the burn in phase and 300 iterations for the
second phase.

```{r f_parent_nlmixr_saem_dfop_tc_1k, results = "hide", warning = FALSE, message = FALSE, dependson = "nlmixr_saem_control"}
f_parent_nlmixr_saem_dfop_tc_1000 <- nlmixr(f_parent_mkin_tc["DFOP", ], est = "saem",
  control = nlmixr_saem_control_1000)
traceplot(f_parent_nlmixr_saem_dfop_tc_1000$nm)
```

Here the fit looks very similar, but we will see below that it shows a higher AIC
than the fit with 800 iterations in the burn in phase. Next we choose
10 000 iterations for the burn in phase and 1000 iterations for the second
phase for comparison with saemix.

```{r f_parent_nlmixr_saem_dfop_tc_10k, results = "hide", warning = FALSE, message = FALSE, dependson = "nlmixr_saem_control"}
f_parent_nlmixr_saem_dfop_tc_10k <- nlmixr(f_parent_mkin_tc["DFOP", ], est = "saem",
  control = nlmixr_saem_control_10k)
traceplot(f_parent_nlmixr_saem_dfop_tc_10k$nm)
```

In the above convergence plot, the time course of 'eta.DMTA_0' and
'log_k2' indicate a false convergence.

The AIC values are internally calculated using Gaussian quadrature.

```{r AIC_parent_nlmixr_saem, cache = FALSE}
AIC(f_parent_nlmixr_saem_sfo_const$nm, f_parent_nlmixr_saem_sfo_tc$nm,
  f_parent_nlmixr_saem_dfop_const$nm, f_parent_nlmixr_saem_dfop_tc$nm,
  f_parent_nlmixr_saem_dfop_tc_1000$nm,
  f_parent_nlmixr_saem_dfop_tc_10k$nm)
```

We can see that again, the DFOP/tc model shows the best goodness of fit.
However, increasing the number of burn-in iterations from 800 to 1000 results
in a higher AIC. If we further increase the number of iterations to 10 000
(burn-in) and 1000 (second phase), the AIC cannot be calculated for the
nlmixr/saem fit, supporting that the fit did not converge properly.

### Comparison

The following table gives the AIC values obtained with the three packages using
the same control parameters (800 iterations burn-in, 300 iterations second
phase, 15 chains).

```{r AIC_all, cache = FALSE}
AIC_all <- data.frame(
  check.names = FALSE,
  "Degradation model" = c("SFO", "SFO", "DFOP", "DFOP"),
  "Error model" = c("const", "tc", "const", "tc"),
  nlme = c(AIC(f_parent_nlme_sfo_const), AIC(f_parent_nlme_sfo_tc), NA, AIC(f_parent_nlme_dfop_tc)),
  nlmixr_focei = sapply(list(f_parent_nlmixr_focei_sfo_const$nm, f_parent_nlmixr_focei_sfo_tc$nm,
  f_parent_nlmixr_focei_dfop_const$nm, f_parent_nlmixr_focei_dfop_tc$nm), AIC),
  saemix = sapply(list(f_parent_saemix_sfo_const$so, f_parent_saemix_sfo_tc$so,
    f_parent_saemix_dfop_const$so, f_parent_saemix_dfop_tc$so), AIC),
  nlmixr_saem = sapply(list(f_parent_nlmixr_saem_sfo_const$nm, f_parent_nlmixr_saem_sfo_tc$nm,
  f_parent_nlmixr_saem_dfop_const$nm, f_parent_nlmixr_saem_dfop_tc$nm), AIC)
)
kable(AIC_all)
```

```{r parms_all, cache = FALSE}
intervals(f_parent_saemix_dfop_tc)
intervals(f_parent_saemix_dfop_tc)
intervals(f_parent_saemix_dfop_tc_10k)
intervals(f_parent_saemix_dfop_tc_mkin_10k)
intervals(f_parent_nlmixr_saem_dfop_tc)
intervals(f_parent_nlmixr_saem_dfop_tc_10k)
```



# References

<!-- vim: set foldmethod=syntax: -->