report_summary.Rmd

---
output:
  word_document:
    fig_caption: true
    fig_height: 10
    fig_width: 10
    reference_docx: bootstrap/initial/data/summaryTemplate.docx
    toc: false
    keep_md: false
params:
  summary_title: ""
  strata: NULL
---


```{r introduction, include = FALSE}

# INTRODUCTION ################################################################

# This markdown documents and integrated scripts analyse biological
# maturity stagings.
# The output is a .docx template that includes
# the results of the analysis and should be used as a standard for
# reporting of maturity staging comparisons.

```

```{r chunk_setup, include=FALSE}
# CHUNK SETUPS #################################################################

knitr::opts_chunk$set(echo = FALSE, warning = FALSE,
                      message=FALSE, results = 'asis', dpi=400)

```


```{r pander_settings, include = FALSE}
# PANDER OPTIONS ##############################################################

panderOptions('table.split.table', Inf)
panderOptions('keep.trailing.zeros', TRUE)
panderOptions('table.alignment.default', "center")
panderOptions('knitr.auto.asis', FALSE)

group <- "adv"

# from extrafont
# font_import(pattern=c("Cali"))
# loadfonts(device="win")
```

---
title: `r config$summary_title`
---

# Executive summary

Text?


*Overview table*

<!-- Stock, Total number of samples (NSample), Preparation method, PA (sex and maturity), CU (sex and maturity), Number of readers (basic and advanced) text -->
```{r overview_statistics_all}
overview_statistics_all <-
    c(Stock=unique(ad_long_all$stock),
    NSample  = length(unique(ad_long_all$SampleID)),
    CU_sex  = cu_tab_sex_all[nrow(cu_tab_sex_all), "Total"],
    CU_mat  = cu_tab_maturity_all[nrow(cu_tab_maturity_all), "Total"],
    PA_sex  = pa_tab_sex_all[nrow(pa_tab_sex_all), "Total"],
    PA_mat  = pa_tab_maturity_all[nrow(pa_tab_maturity_all), "Total"],
    N_basic=length(unique(ad_long_all$reader))-length(unique(ad_long_adv$reader)),
    N_adv=length(unique(ad_long_adv$reader)))

# Table caption
set.caption('**Table X:** Overview table: Stock; Total number of samples (NSample); coefficient of unalikeability (CU) for sex (CU_sex) and maturity (CU_maturity); percentage of agreement (PA) for sex (PA_sex) and for maturity (PA_maturity); number of basic readers (N_basic) and number of advanced readers (N_adv).')

# Output table
pander(style_table0(overview_statistics_all), missing="")

```

# Overview of samples and advanced stagers

Text?



\br

```{r sample_overview}

# Table caption
set.caption(
  paste(
    "**Table X:** Overview of the number of samples (num) by length category used for the exchange event number ",
    config$event_id,
    ". The range of maturity stages (mat_stages), modal maturity stages (modal_mat_stages), sex categories (sex_cat) and modal sex categories (modal_sex_cat) for those individuals within the length range are also presented.",
    sep = ""
  )
)

# Output table
pander(style_table0(sample_data_overview), style = "simple")

```

\br

```{r participants_overview}

# PARTICIPATANTS OVERVIEW #####################################################

# Table caption
set.caption('**Table X:** Overview of maturity stagers.')

# Output table
pander(style_table0(filter(stager_data, Expertise!="Basic")))
```


<!-- ## Advanced stagers -->

<!-- **All samples included**  -->

# Results overview

**Multimodal cases**

If there were available histological samples for all the fish individuals in the exchange event, no multiple modes are expected, however, if there were no hitology samples for some of them, multiple maturity stage modes could be found. In the next table this information is presented for the sex category determination. 

<!-- Total number of samples (NSample), number (CountMM) and percentage of multimodal cases (PercMM) text -->
```{r summary_multiple_modes_sex}

histN=ad %>% subset(DoesSampleHaveHistologyImage=="Yes") %>% select(FishID) %>% unique() %>% dim()
allN=ad_long_all %>% select(FishID) %>% unique() %>% dim()
perc_not_Hist=100-round(100*histN[1]/allN[1], 0)

summary_multiple_modes_sex_all <-
  c(NSample  = length(unique(ad_long_all$FishID)),
    Nhist=histN[1],
    Perc_not_Hist=perc_not_Hist,
    PercMM_traditional=unique(ifelse(multimode_cases_tab_traditional_Sex_all$NModes_trad=="zero", paste(0,"%"), paste(round((nrow(multimode_cases_tab_traditional_Sex_all)/length(unique(ad_long_all$SampleID)))*100, digits=0),"%"))),
    PercMM_linear_weight=unique(ifelse(multimode_cases_tab_linear_Sex_all$NModes_linear=="zero", paste(0,"%"), paste(round((nrow(multimode_cases_tab_linear_Sex_all)/length(unique(ad_long_all$SampleID)))*100, digits=0),"%"))),
    PercMM_negexp_weight=unique(ifelse(multimode_cases_tab_negexp_Sex_all$NModes_negexp=="zero", paste(0,"%"), paste(round((nrow(multimode_cases_tab_negexp_Sex_all)/length(unique(ad_long_all$SampleID)))*100, digits=0),"%"))),
    PercMM_multistage=unique(ifelse(multimode_cases_tab_multistage_Sex_all$NModes_multistage=="zero", paste(0,"%"), paste(round((nrow(multimode_cases_tab_multistage_Sex_all)/length(unique(ad_long_all$SampleID)))*100, digits=0),"%"))))

# Table caption
set.caption('**Table X:** Summary of statistics for sex staging; Total number of fish individuals studied (NSample), number of fish individuals with histological samples (Nhist), percentage of fish individuals without histology (Perc_not_Hist). The percentage of cases (fish samples) with multiple modes depending on the approach to weight the experience of the stager which will be considered when defining the fish sex stage mode. PercMM_traditional shows the percentage of the total samples for which multiple modes are obtained when all the stagers are equally weighted. PercMM_linear_weight shows the percentage of the total samples for which multiple modes are obtained when the weight assigned to the different stagers decreases linearly with the experience, while in the PercMM_negexp the weight applied decreases with a negative exponential shape with the experience. The PercMM_multistage shows the percentage of multiple mode cases when a combination of the different methodologies is used, as explained in the material and methods section')

# Output table
pander(style_table0(summary_multiple_modes_sex_all), missing="")


```


If there were available histological samples for all the fish individuals in the exchange event, no multiple modes are expected, however, if there were no hitology samples for some of them, multiple maturity stage modes could be found. In the next table this information is presented for the maturity staging. 

<!-- Total number of samples (NSample), number (CountMM) and percentage of multimodal cases (PercMM) text -->
```{r summary_multiple_modes_maturity}

if(!isTRUE(config$maturity_hist)){

histN=ad %>% subset(DoesSampleHaveHistologyImage=="Yes") %>% select(FishID) %>% unique() %>% dim()
allN=ad_long_all %>% select(FishID) %>% unique() %>% dim()
perc_not_Hist=100-round(100*histN[1]/allN[1], 0)

summary_multiple_modes_maturity_all <-
  c(NSample  = length(unique(ad_long_all$FishID)),
    Nhist=histN[1],
    Perc_not_Hist=perc_not_Hist,
    PercMM_traditional=unique(ifelse(multimode_cases_tab_traditional_Maturity_all$NModes_trad=="zero", paste(0,"%"), paste(round((nrow(multimode_cases_tab_traditional_Maturity_all)/length(unique(ad_long_all$SampleID)))*100, digits=0),"%"))),
    PercMM_linear_weight=unique(ifelse(multimode_cases_tab_linear_Maturity_all$NModes_linear=="zero", paste(0,"%"), paste(round((nrow(multimode_cases_tab_linear_Maturity_all)/length(unique(ad_long_all$SampleID)))*100, digits=0),"%"))),
    PercMM_negexp_weight=unique(ifelse(multimode_cases_tab_negexp_Maturity_all$NModes_negexp=="zero", paste(0,"%"), paste(round((nrow(multimode_cases_tab_negexp_Maturity_all)/length(unique(ad_long_all$SampleID)))*100, digits=0),"%"))),
    PercMM_multistage=unique(ifelse(multimode_cases_tab_multistage_Maturity_all$NModes_multistage=="zero", paste(0,"%"), paste(round((nrow(multimode_cases_tab_multistage_Maturity_all)/length(unique(ad_long_all$SampleID)))*100, digits=0),"%"))))

# Table caption
set.caption('**Table X:** Summary of statistics for maturity staging; Total number of fish individuals studied (NSample), number of fish individuals with histological samples (Nhist), percentage of fish individuals without histology (Perc_not_Hist). The percentage of cases (fish samples) with multiple modes depending on the approach to weight the experience of the stager which will be considered when defining the fish maturity stage mode. PercMM_traditional shows the percentage of the total samples for which multiple modes are obtained when all the stagers are equally weighted. PercMM_linear_weight shows the percentage of the total samples for which multiple modes are obtained when the weight assigned to the different stagers decreases linearly with the experience, while in the PercMM_negexp the weight applied decreases with a negative exponential shape with the experience. The PercMM_multistage shows the percentage of multiple mode cases when a combination of the different methodologies is used, as explained in the material and methods section')

# Output table
pander(style_table0(summary_multiple_modes_maturity_all), missing="")

}

```


**Sex categorization table**

```{r sex_composition_tab_adv}

# Sex categorization TABLE - Advanced stagers #################################################

# Table caption
set.caption('**Table X:** Sex categorization table: presents the
             number of categorizations made per expert for each modal sex category.')

# Output table
pander(style_table0(sex_composition_tab_adv), missing = "-", style = "simple")

```


**Maturity staging table**

```{r num_read_maturity_adv}

# maturity stagings TABLE - Advanced stagers #################################################

# Table caption
set.caption('**Table X:** Maturity staging table: presents the
             number of stagings made per expert stager for each modal maturity stage.')

# Output table
pander(style_table0(maturity_composition_tab_adv), missing = "-", style = "simple")

```


**Coefficient of Unalikeability (CU) table by modal sex category**

```{r cv_sex_adv}

# CU TABLE by modal sex category - Advanced stagers #################################################

# Table caption
set.caption('**Table X:** Coefficient of unlikeability (CU) table by modal sex category: presents the
             CU per modal sex category and advanced stager, the CU of all advanced
             stagers combined per sex category stage and a weighted mean of the CU
             per stager.')

# Output table
pander(style_table1(cu_tab_sex_adv), missing = "-")

```


**Coefficient of Unalikeability (CU) table by modal maturity stage**

```{r cv_maturity_adv}

# CU TABLE by modal maturity stage - Advanced stagers #################################################

# Table caption
set.caption('**Table X:** Coefficient of unlikeability (CU) table by modal maturity stage: presents the
             CU per modal maturity and advanced stager, the CU of all advanced
             stagers combined per modal maturity stage and a weighted mean of the CU
             per stager.')

# Output table
pander(style_table1(cu_tab_maturity_adv), missing = "-")

```


**PA table by modal sex category**

```{r percentage_agreement_sex_adv}

# PERCENTAGE AGREEMENT TABLE by modal sex category - Advanced stagers ###############################

# Table caption
set.caption('**Table X:** Percentage agreement (PA) table by modal sex category: presents the PA per
             modal sex category and stager, of all advanced stagers
             combined and a weighted mean of the PA per stager.')

# Output table
pander(style_table1(pa_tab_sex_adv), missing = "-")

```


**PA table by modal maturity stage**

```{r percentage_agreement_maturity_adv}

# PERCENTAGE AGREEMENT TABLE by modal maturity stage - Advanced stagers ###############################

# Table caption
set.caption('**Table X:** Percentage agreement (PA) table: presents the PA per
             modal maturity and stager, of all advanced stagers
             combined and a weighted mean of the PA per stager.')

# Output table
pander(style_table1(pa_tab_maturity_adv), missing = "-")

```



**General frequency bias table by modal sex category **

```{r rb_sex_adv,results='asis'}

# Frequency bias table by modal sex category - Advanced stagers ######################################

# Table caption
set.caption('**Table X:** Frequency bias table by modal sex category: presents the frequency of the different sex categories per
             modal maturity for all the advanced stager together.')

# Output table
pander(style_table0(general_bias_freq_tab_sex_adv), missing = "-")

```


**General frequency bias table by modal maturity stage**

```{r rb_maturity_adv,results='asis'}

# Frequency bias table by modal maturity stage - Advanced stagers ######################################

# Table caption
set.caption('**Table X:** Frequency bias table by modal maturity stage: presents the frequency of the different maturity stages per
             modal maturity stage for all the advanced stager together.')

# Output table
pander(style_table0(general_bias_freq_tab_maturity_adv), missing = "-")

```


**General Frequency Bias plot by modal sex category**

```{r bias_plots_sex_exp, fig.width = 5, fig.height = 4, fig.cap = cap_in}

# BIAS PLOT by modal sex category - Advanced stagers ################################################

# Table caption
cap_in <- '**Figure X:** Sex categorization bias plot by modal sex category for advanced stagers: presents the frequency per
             modal  sex category and sex category for all the advanced stager together.  '

# Output figure
plot_general_freq_sex(ad_long_adv, strata=NULL)


```


**General Frequency Bias plot by modal maturity stage**

```{r bias_plots_maturity_exp, fig.width = 5, fig.height = 4, fig.cap = cap_in}

# BIAS PLOT by modal maturity stage - Advanced stagers ################################################

# Table caption
cap_in <- '**Figure X:** Maturity staging bias plot by modal maturity stage for advanced stagers:  presents the frequency per
             modal maturity and maturity stage for all the advanced stager together.  '

# Output figure
plot_general_freq_matur(ad_long_adv, strata=NULL)

```



```{r set_strata}
# initialise strata loop
istrata <- 0
print_strata <- function() length(config$strata) >= istrata
```

<!-- first strata -->

```{r strata_plus_one}
# second strata
istrata <- istrata + 1
```

```{r strata_title, eval = print_strata()}
stratum <- config$strata[istrata]
# Section title
asis_output(paste("# Results by", stratum))
```


```{r ar_sex_title, eval = print_strata()}
#  title
asis_output(paste0("**Sex categorization by ", stratum, "**"))
```

```{r ar_sex_by, eval = print_strata()}
# NUMBER OF SEX CATEGORIZATION PER STOCK - Advanced stagers #########################

# Table caption
set.caption(paste0('**Table X:** Number of sex categorization per ', stratum, ' for advanced stagers.'))

# Output table
pander(style_table1(get(vsname("num_read_tab_modal_sex_by"))), missing = "-")
```


```{r ar_matur_title, eval = print_strata()}
#  title
asis_output(paste0("**Maturity staging by ", stratum, "**"))
```

```{r ar_matur_by, eval = print_strata()}
# NUMBER OF MATURITY STAGING PER STOCK - Advanced stagers #########################

# Table caption
set.caption(paste0('**Table X:** Number of maturity stagings per ', stratum, ' for advanced stagers.'))

# Output table
pander(style_table1(get(vsname("num_read_tab_modal_matur_by"))), missing = "-")
```



```{r cu_sex_title, eval = print_strata()}
#  title
asis_output(paste0("**Coefficient of Unalikeability by modal sex category ", stratum, "**"))
```


```{r cu_sex_by, eval = print_strata()}

# Table caption
set.caption(paste0('**Table X:** CU by modal sex category per ', stratum, '.'))

# Output table
pander(style_table2(get(vsname("cu_tab_sex_by"))), missing = "-")

```



```{r cu_matur_title, eval = print_strata()}
#  title
asis_output(paste0("**Coefficient of Unalikeability by modal maturity stage ", stratum, "**"))
```


```{r cu_matur_by, eval = print_strata()}

# Table caption
set.caption(paste0('**Table X:** CU by modal maturity stage per ', stratum, '.'))

# Output table
pander(style_table2(get(vsname("cu_tab_maturity_by"))), missing = "-")

```


```{r pa_sex_title, eval = print_strata()}
#  title
asis_output(paste0("**Percentage Agreement by sex category per ", stratum, "**"))
```

```{r pa_sex_by, eval = print_strata()}
# Section title
#asis_("## PA  by sex category - Advanced stagers")  #    #  #############################################

# Table caption
set.caption(paste0('**Table X:** Percentage Agreement  by sex category per ', stratum, '.'))

# Output table
pander(style_table2(get(vsname("pa_tab_sex_by"))), missing = "-")

```



```{r pa_matur_title, eval = print_strata()}
#  title
asis_output(paste0("**Percentage Agreement by maturation stage per ", stratum, "**"))
```

```{r pa_matur_by, eval = print_strata()}
# Section title
#asis_("## PA by maturation stage - Advanced stagers")  #    #  #############################################

# Table caption
set.caption(paste0('**Table X:** Percentage Agreement by maturation stage per ', stratum, '.'))

# Output table
pander(style_table2(get(vsname("pa_tab_maturity_by"))), missing = "-")

```


<!-- second strata -->

```{r ref.label='strata_plus_one'}
```

```{r ref.label='strata_title', eval = print_strata()}
```

```{r ref.label='ar_sex_title', eval = print_strata()}
```

```{r ref.label='ar_sex_by', eval = print_strata()}
```

```{r ref.label='ar_matur_title', eval = print_strata()}
```

```{r ref.label='ar_matur_by', eval = print_strata()}
```

```{r ref.label='cu_sex_title', eval = print_strata()}
```

```{r ref.label='cu_sex_by', eval = print_strata()}
```

```{r ref.label='cu_matur_title', eval = print_strata()}
```

```{r ref.label='cu_matur_by', eval = print_strata()}
```

```{r ref.label='pa_sex_title', eval = print_strata()}
```

```{r ref.label='pa_sex_by', eval = print_strata()}
```

```{r ref.label='pa_matur_title', eval = print_strata()}
```

```{r ref.label='pa_matur_by', eval = print_strata()}
```


<!-- third strata -->

```{r ref.label='strata_plus_one'}
```

```{r ref.label='strata_title', eval = print_strata()}
```

```{r ref.label='ar_sex_title', eval = print_strata()}
```

```{r ref.label='ar_sex_by', eval = print_strata()}
```

```{r ref.label='ar_matur_title', eval = print_strata()}
```

```{r ref.label='ar_matur_by', eval = print_strata()}
```

```{r ref.label='cu_sex_title', eval = print_strata()}
```

```{r ref.label='cu_sex_by', eval = print_strata()}
```

```{r ref.label='cu_matur_title', eval = print_strata()}
```

```{r ref.label='cu_matur_by', eval = print_strata()}
```

```{r ref.label='pa_sex_title', eval = print_strata()}
```

```{r ref.label='pa_sex_by', eval = print_strata()}
```

```{r ref.label='pa_matur_title', eval = print_strata()}
```

```{r ref.label='pa_matur_by', eval = print_strata()}
```


<!-- fourth strata -->

```{r ref.label='strata_plus_one'}
```

```{r ref.label='strata_title', eval = print_strata()}
```

```{r ref.label='ar_sex_title', eval = print_strata()}
```

```{r ref.label='ar_sex_by', eval = print_strata()}
```

```{r ref.label='ar_matur_title', eval = print_strata()}
```

```{r ref.label='ar_matur_by', eval = print_strata()}
```

```{r ref.label='cu_sex_title', eval = print_strata()}
```

```{r ref.label='cu_sex_by', eval = print_strata()}
```

```{r ref.label='cu_matur_title', eval = print_strata()}
```

```{r ref.label='cu_matur_by', eval = print_strata()}
```

```{r ref.label='pa_sex_title', eval = print_strata()}
```

```{r ref.label='pa_sex_by', eval = print_strata()}
```

```{r ref.label='pa_matur_title', eval = print_strata()}
```

```{r ref.label='pa_matur_by', eval = print_strata()}
```


# Conclusion