In ready4-dev/TTU: Implement Transfer to Utility Mapping Algorithms
\blandscape
This workflow has been deprecated. Many of the functions included in this workflow have been replaced by methods in the specific package. Updated syntax for reporting utility mapping studies with the TTU package is currently being developed. The current syntax should be regarded as unstable and subject to change at short notice.
This vignette is an example of implementing of a reproducible workflow to:
-
develop transfer to utility models to predict AQoL-6D in a clinical mental health sample;
-
create and share catalogues of these models; and
-
author a scientific summary of the results and methods deployed.
This example, should be read in conjunction with the vignette illustrating a reporting workflow implemented for developing an EQ-5D TTU using alternative predictors, which highlights some additional considerations. It is highly recommended that you only implement this reporting workflow for your own projects once you have identified an optimal set of input parameters following an initial exploratory analysis. The benefits of undertaking such initial exploratory work include identifying where it may be appropriate to override algorithm default selections for preferred model types and covariates and discovering if it is necessary to supply non-default input parameters to address problems in model convergence.
As with all the vignettes included with the TTU package, this example is illustrated using entirely fake data, so do not use its outputs to inform decision making. For an example of how TTU workflows were applied to real data see https://www.medrxiv.org/content/10.1101/2021.07.07.21260129v2.full .
library(TTU)
library(specific)
Input parameters
Reporting parameters
Bibliograhic data
We begin by specifying the authorship and other bibliographic data that we wish to attach to all reports that we generate. We use the ready4show_authors and ready4show_institutes classes of the ready4show package to do this.
authors_tb <- ready4show::ready4show_authors() %>%
tibble::add_case(first_nm_chr = "Alejandra",
middle_nm_chr = "Rocio",
last_nm_chr = "Scienceace",
title_chr = "Dr",
qualifications_chr = "MD, PhD",
institute_chr = "Institute_A, Institute_B",
sequence_int = 1,
is_corresponding_lgl = T,
email_chr = "fake_email@fake_institute.com") %>%
tibble::add_case(first_nm_chr = "Fionn",
middle_nm_chr = "Seamus",
last_nm_chr = "Researchchamp",
title_chr = "Prof",
qualifications_chr = "MSc, PhD",
institute_chr = "Institute_C, Institute_B",
sequence_int = 2,
email_chr = "fake_email@fake_institute.com")
institutes_tb <- ready4show::ready4show_institutes() %>%
tibble::add_case(short_name_chr = "Institute_A", long_name_chr = "Awesome University, Shanghai") %>%
tibble::add_case(short_name_chr = "Institute_B", long_name_chr = "August Institution, London") %>%
tibble::add_case(short_name_chr = "Institute_C", long_name_chr = "Highly Ranked Uni, Montreal")
We combine the data about authors with other bibliographic information.
header_yaml_args_ls <- ready4show::make_header_yaml_args_ls(authors_tb = authors_tb,
institutes_tb = institutes_tb,
title_1L_chr = "A hypothetical study using fake data for instructional purposes only",
keywords_chr = c("this","is","a","replication","using","fake","data","do", "not","cite"))
Report formatting
We add parameters relating to the desired output format of the reports we will be producing.
output_format_ls <- ready4show::make_output_format_ls(manuscript_outp_1L_chr = "PDF",
manuscript_digits_1L_int = 2L,
supplementary_outp_1L_chr = "PDF",
supplementary_digits_1L_int = 2L)
Data parameters
Dataset
We read in our study dataset (in this case we are a fake dataset).
ds_tb <- youthvars::replication_popl_tb %>%
youthvars::transform_raw_ds_for_analysis()
The first few records of the dataset are as follows.
ds_tb %>%
head() %>%
ready4show::print_table(output_type_1L_chr = "HTML",
caption_1L_chr = "Input dataset",
mkdn_tbl_ref_1L_chr = "tab:inputds")
Data dictionary
A data dictionary must also be supplied for the dataset. The dictionary must be of class ready4use_dictionary from the ready4use package. In this case, we are using a pre-existing dictionary.
dictionary_tb <- youthvars::make_final_repln_ds_dict()
The first few records of the data dictionary are reproduced below.
dictionary_tb %>%
head() %>%
ready4show::print_table(output_type_1L_chr = "HTML",
caption_1L_chr = "Data dictionary",
mkdn_tbl_ref_1L_chr = "tab:dictionary")
Dataset metadata
We describe the features of our dataset that are most relevant to the analyses we are planning to undertake. Note all, Multi-Attribute Utility Instrument (MAUI) question variables must start with a common prefix, which needs to be passed to the maui_item_pfx_1L_chr argument. The dataset does not need to include variables for total health utility scores (weighted and unweighted), as these variables will be computed during the analysis. However, you can supply the names you wish assigned to these variables once they are computed using the utl_wtd_var_nm_1L_chr and utl_unwtd_var_nm_1L_chr arguments.
ds_descvs_ls <- make_ds_descvs_ls(candidate_predrs_chr = c("K6","PHQ9"),#
candidate_covar_nms_chr = c("d_age", "SOFAS",
"c_p_diag_s",
"c_clinical_staging_s",
"d_relation_s",
"d_studying_working"),
cohort_descv_var_nms_chr = c("d_age","d_relation_s",
"d_studying_working",
"c_p_diag_s","c_clinical_staging_s",
"SOFAS"),
dictionary_tb = dictionary_tb,
id_var_nm_1L_chr = "fkClientID",
is_fake_1L_lgl = T,
msrmnt_date_var_nm_1L_chr = "d_interview_date",
round_var_nm_1L_chr = "round",
round_vals_chr = c("Baseline", "Follow-up"),
maui_item_pfx_1L_chr = "aqol6d_q",
utl_wtd_var_nm_1L_chr = "aqol6d_total_w",
utl_unwtd_var_nm_1L_chr = "aqol6d_total_c")
Candidate predictors metadata
We create a lookup table with further metadata on the dataset variables that we specified as our candidate predictors in the candidate_predrs_chr element of ds_descvs_ls, as well as the preferred covariates that we will specify later in the make_input_params function.
predictors_lup <- specific::specific_predictors(specific::make_pt_specific_predictors(short_name_chr = c(ds_descvs_ls$candidate_predrs_chr, "SOFAS"),
long_name_chr = c("K6 total score", "PHQ9 total score", "SOFAS total score"),
min_val_dbl = 0,
max_val_dbl = c(24,27,100),
class_chr = "integer",
increment_dbl = 1,
class_fn_chr = c("youthvars::youthvars_k6",
"youthvars::youthvars_phq9",
"youthvars::youthvars_sofas"),
mdl_scaling_dbl = 0.01,
covariate_lgl = F))
Multi-Attribute Utility Instrument (MAUI) parameters
We also need to provide information specific to the MAUI used to collect the data from which health utility will be derived. The minimum allowable health utility weight for this instrument can be declared using the utl_min_val_1L_dbl argument.
maui_params_ls <- make_maui_params_ls(maui_domains_pfxs_1L_chr = "vD",
maui_itm_short_nms_chr = c("Household tasks",
"Getting around","Morbility",
"Self care","Enjoy close rels",
"Family rels",
"Community involvement", "Despair",
"Worry", "Sad", "Agitated",
"Energy level","Control",
"Coping","Frequency of pain",
"Degree of pain",
"Pain interference", "Vision",
"Hearing", "Communication"),
maui_scoring_fn = youthvars::add_adol6d_scores,
short_and_long_nm = c("AQoL-6D",
"Assessment of Quality of Life - Six Dimension"),
utl_min_val_1L_dbl = 0.03)
Analysis parameters
We can also (but do not have to) specify how we want to adapt the input parameters we have just declared for any secondary analyses (e.g. that use different combinations of candidate predictors and covariates). We can do this using the make_scndry_anlys_params function.
scndry_anlys_params_ls <- make_scndry_anlys_params(candidate_predrs_chr = c("SOFAS"),
prefd_covars_chr = NA_character_)
We can now create a list of input parameters using the make_input_params function, principally using the objects created in the preceding steps. We can force the use of specified model types or covariates (rather than allowing the analysis algorithm to select these based on automated criteria) by using the prefd_mdl_types_chr and prefd_covars_chr arguments. As we have specified secondary analysis parameters, we supply these to the scndry_anlys_params_ls argument.
If during set-up, you have created an empty dataverse dataset to post results to, assign the DOI for your dataset and the name of the dataverse containing it to the dv_ds_nm_and_url_chr argument (your values will be different to those below, which will not work for you). If you do not wish to post results to a dataverse dataset, set dv_ds_nm_and_url_chr = NULL. The make_input_params function performs a number of checks relating to the naming conventions used in some variables of interest and where necessary (for consistency with reporting templates) modifies the names of these variables.
input_params_ls <- make_input_params(ds_tb,
ds_descvs_ls = ds_descvs_ls,
dv_ds_nm_and_url_chr = c("fakes", # Dataverse containing dataset
"https://doi.org/10.7910/DVN/D74QMP"), # DOI
header_yaml_args_ls = header_yaml_args_ls,
maui_params_ls = maui_params_ls,
output_format_ls = output_format_ls,
predictors_lup = predictors_lup,
prefd_covars_chr = "SOFAS",
prefd_mdl_types_chr = c("OLS_CLL","GLM_GSN_LOG"),
scndry_anlys_params_ls = scndry_anlys_params_ls)
Analyse, Report and Share
Run analysis
All analytic steps for our primary and secondary analyses are executed with the following function call. Note, this step can involve long execution times (potentially over an hour, depending on the input parameters).
write_analyses(input_params_ls)
Report results
We now generate the catalogues (one for the primary and one for each secondary analysis) that report the models created in the preceding step. The write_mdl_smry_rprt function also creates shareable versions (i.e. with copies of source dataset replaced by synthetic data) of the models. This step can take several minutes to execute.
input_params_ls <- write_mdl_smry_rprt(input_params_ls,
use_shareable_mdls_1L_lgl = T)
Share results
We can now share the results created in the previous step by posting them to the online repository, the details of which we supplied earlier.
write_study_outp_ds(input_params_ls)
The write_study_outp_ds function writes its outputs to a draft dataset - which you can review by logging into your dataverse account, before clicking on the option to publish the dataset. The stucture of your draft dataset should be similar to those produced by this vignette, which are viewable at: https://doi.org/10.7910/DVN/D74QMP
Create manuscript
Create study metadata
The first step to creating a manuscript that provides the scientific summary of the study is to add some metadata about the study itself to our parameters list, using the make_study_descs_ls function. If you want the manuscript to refer to some of the candidate predictors by different names than that provided in the short_name_chr column of the predictors_lup object defined earlier, you can create a table of name correspondences and pass it to the var_nm_change_lup argument.
input_params_ls <- make_study_descs_ls(input_params_ls = input_params_ls,
background_1L_chr = "Our study is entirely fictional and has been created to illustrate TTU package functionality.",
coi_1L_chr = "None declared.",
conclusion_1L_chr = "If this study was real, the results would be interesting.",
ethics_1L_chr = "The study was reviewed and granted approval by Awesome University's Human Research Ethics Committee (1111111.1).",
funding_1L_chr = "The study was funded by Generous Benefactor.",
sample_desc_1L_chr = "The study sample is fake data that pretends to be young people aged 12 to 25 years who attended Australian primary care services for mental health related needs between November 2019 to August 2020.",
time_btwn_bl_and_fup_1L_chr = "three months",
var_nm_change_lup = tibble::tibble(old_nms_chr = c("PHQ9","GAD7"),
new_nms_chr = c("PHQ-9",
"GAD-7")))
Render auto-generated first-draft
We can now pass the updated input parameters list to the write_manuscript function. This will produce an initial draft of a scientific manuscript (detailed methods and results, a basic introduction and no discussion), by first downloading an extension to the TTU package (https://github.com/ready4-dev/ttu_lng_ss). If (AND ONLY IF), you are sure that you will never develop the manuscript further with a view to future publication, you can add this automatically generated draft to the study dataset by adding write_to_dv_1L_lgl = T to the function call (as below).
results_ls <- ready4show::write_manuscript(input_params_ls = input_params_ls,
write_to_dv_1L_lgl = T)
The manuscript exported to the dataset by the preceding call is accessible at: https://dataverse.harvard.edu/api/access/datafile/4957407 . A local version will also automatically be saved on your machine and you can discover where this is by printing the object results_ls$path_params_ls$paths_ls$reports_dir_1L_chr.
You can override the output format specified in the input parameter list object by using the output_type_1L_chr argument. Note, that this time around we are using the results_ls object as the main input as this object was automatically saved (to the directory path stored in results_ls$path_params_ls$paths_ls$output_data_dir_1L_chr) by the preceding call.
Modify for publication
If you wish to develop the manuscript into something you will submit for publication, consult https://github.com/ready4-dev/ttu_lng_ss for guidance.
Purge dataset copies
As a final step, it is important to delete all local copies (but not the original version) of the study dataset that were produced during the analysis.
write_to_delete_ds_copies(results_ls)
\elandscape
ready4-dev/TTU documentation built on July 2, 2024, 8:12 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
\blandscape This workflow has been deprecated. Many of the functions included in this workflow have been replaced by methods in the specific package. Updated syntax for reporting utility mapping studies with the TTU package is currently being developed. The current syntax should be regarded as unstable and subject to change at short notice.
This vignette is an example of implementing of a reproducible workflow to:
-
develop transfer to utility models to predict AQoL-6D in a clinical mental health sample;
-
create and share catalogues of these models; and
-
author a scientific summary of the results and methods deployed.
This example, should be read in conjunction with the vignette illustrating a reporting workflow implemented for developing an EQ-5D TTU using alternative predictors, which highlights some additional considerations. It is highly recommended that you only implement this reporting workflow for your own projects once you have identified an optimal set of input parameters following an initial exploratory analysis. The benefits of undertaking such initial exploratory work include identifying where it may be appropriate to override algorithm default selections for preferred model types and covariates and discovering if it is necessary to supply non-default input parameters to address problems in model convergence.
As with all the vignettes included with the TTU package, this example is illustrated using entirely fake data, so do not use its outputs to inform decision making. For an example of how TTU workflows were applied to real data see https://www.medrxiv.org/content/10.1101/2021.07.07.21260129v2.full .
library(TTU) library(specific)
Input parameters
Reporting parameters
Bibliograhic data
We begin by specifying the authorship and other bibliographic data that we wish to attach to all reports that we generate. We use the ready4show_authors and ready4show_institutes classes of the ready4show package to do this.
authors_tb <- ready4show::ready4show_authors() %>% tibble::add_case(first_nm_chr = "Alejandra", middle_nm_chr = "Rocio", last_nm_chr = "Scienceace", title_chr = "Dr", qualifications_chr = "MD, PhD", institute_chr = "Institute_A, Institute_B", sequence_int = 1, is_corresponding_lgl = T, email_chr = "fake_email@fake_institute.com") %>% tibble::add_case(first_nm_chr = "Fionn", middle_nm_chr = "Seamus", last_nm_chr = "Researchchamp", title_chr = "Prof", qualifications_chr = "MSc, PhD", institute_chr = "Institute_C, Institute_B", sequence_int = 2, email_chr = "fake_email@fake_institute.com")
institutes_tb <- ready4show::ready4show_institutes() %>% tibble::add_case(short_name_chr = "Institute_A", long_name_chr = "Awesome University, Shanghai") %>% tibble::add_case(short_name_chr = "Institute_B", long_name_chr = "August Institution, London") %>% tibble::add_case(short_name_chr = "Institute_C", long_name_chr = "Highly Ranked Uni, Montreal")
We combine the data about authors with other bibliographic information.
header_yaml_args_ls <- ready4show::make_header_yaml_args_ls(authors_tb = authors_tb, institutes_tb = institutes_tb, title_1L_chr = "A hypothetical study using fake data for instructional purposes only", keywords_chr = c("this","is","a","replication","using","fake","data","do", "not","cite"))
Report formatting
We add parameters relating to the desired output format of the reports we will be producing.
output_format_ls <- ready4show::make_output_format_ls(manuscript_outp_1L_chr = "PDF", manuscript_digits_1L_int = 2L, supplementary_outp_1L_chr = "PDF", supplementary_digits_1L_int = 2L)
Data parameters
Dataset
We read in our study dataset (in this case we are a fake dataset).
ds_tb <- youthvars::replication_popl_tb %>% youthvars::transform_raw_ds_for_analysis()
The first few records of the dataset are as follows.
ds_tb %>% head() %>% ready4show::print_table(output_type_1L_chr = "HTML", caption_1L_chr = "Input dataset", mkdn_tbl_ref_1L_chr = "tab:inputds")
Data dictionary
A data dictionary must also be supplied for the dataset. The dictionary must be of class ready4use_dictionary from the ready4use package. In this case, we are using a pre-existing dictionary.
dictionary_tb <- youthvars::make_final_repln_ds_dict()
The first few records of the data dictionary are reproduced below.
dictionary_tb %>% head() %>% ready4show::print_table(output_type_1L_chr = "HTML", caption_1L_chr = "Data dictionary", mkdn_tbl_ref_1L_chr = "tab:dictionary")
Dataset metadata
We describe the features of our dataset that are most relevant to the analyses we are planning to undertake. Note all, Multi-Attribute Utility Instrument (MAUI) question variables must start with a common prefix, which needs to be passed to the maui_item_pfx_1L_chr argument. The dataset does not need to include variables for total health utility scores (weighted and unweighted), as these variables will be computed during the analysis. However, you can supply the names you wish assigned to these variables once they are computed using the utl_wtd_var_nm_1L_chr and utl_unwtd_var_nm_1L_chr arguments.
ds_descvs_ls <- make_ds_descvs_ls(candidate_predrs_chr = c("K6","PHQ9"),# candidate_covar_nms_chr = c("d_age", "SOFAS", "c_p_diag_s", "c_clinical_staging_s", "d_relation_s", "d_studying_working"), cohort_descv_var_nms_chr = c("d_age","d_relation_s", "d_studying_working", "c_p_diag_s","c_clinical_staging_s", "SOFAS"), dictionary_tb = dictionary_tb, id_var_nm_1L_chr = "fkClientID", is_fake_1L_lgl = T, msrmnt_date_var_nm_1L_chr = "d_interview_date", round_var_nm_1L_chr = "round", round_vals_chr = c("Baseline", "Follow-up"), maui_item_pfx_1L_chr = "aqol6d_q", utl_wtd_var_nm_1L_chr = "aqol6d_total_w", utl_unwtd_var_nm_1L_chr = "aqol6d_total_c")
Candidate predictors metadata
We create a lookup table with further metadata on the dataset variables that we specified as our candidate predictors in the candidate_predrs_chr element of ds_descvs_ls, as well as the preferred covariates that we will specify later in the make_input_params function.
predictors_lup <- specific::specific_predictors(specific::make_pt_specific_predictors(short_name_chr = c(ds_descvs_ls$candidate_predrs_chr, "SOFAS"), long_name_chr = c("K6 total score", "PHQ9 total score", "SOFAS total score"), min_val_dbl = 0, max_val_dbl = c(24,27,100), class_chr = "integer", increment_dbl = 1, class_fn_chr = c("youthvars::youthvars_k6", "youthvars::youthvars_phq9", "youthvars::youthvars_sofas"), mdl_scaling_dbl = 0.01, covariate_lgl = F))
Multi-Attribute Utility Instrument (MAUI) parameters
We also need to provide information specific to the MAUI used to collect the data from which health utility will be derived. The minimum allowable health utility weight for this instrument can be declared using the utl_min_val_1L_dbl argument.
maui_params_ls <- make_maui_params_ls(maui_domains_pfxs_1L_chr = "vD", maui_itm_short_nms_chr = c("Household tasks", "Getting around","Morbility", "Self care","Enjoy close rels", "Family rels", "Community involvement", "Despair", "Worry", "Sad", "Agitated", "Energy level","Control", "Coping","Frequency of pain", "Degree of pain", "Pain interference", "Vision", "Hearing", "Communication"), maui_scoring_fn = youthvars::add_adol6d_scores, short_and_long_nm = c("AQoL-6D", "Assessment of Quality of Life - Six Dimension"), utl_min_val_1L_dbl = 0.03)
Analysis parameters
We can also (but do not have to) specify how we want to adapt the input parameters we have just declared for any secondary analyses (e.g. that use different combinations of candidate predictors and covariates). We can do this using the make_scndry_anlys_params function.
scndry_anlys_params_ls <- make_scndry_anlys_params(candidate_predrs_chr = c("SOFAS"), prefd_covars_chr = NA_character_)
We can now create a list of input parameters using the make_input_params function, principally using the objects created in the preceding steps. We can force the use of specified model types or covariates (rather than allowing the analysis algorithm to select these based on automated criteria) by using the prefd_mdl_types_chr and prefd_covars_chr arguments. As we have specified secondary analysis parameters, we supply these to the scndry_anlys_params_ls argument.
If during set-up, you have created an empty dataverse dataset to post results to, assign the DOI for your dataset and the name of the dataverse containing it to the dv_ds_nm_and_url_chr argument (your values will be different to those below, which will not work for you). If you do not wish to post results to a dataverse dataset, set dv_ds_nm_and_url_chr = NULL. The make_input_params function performs a number of checks relating to the naming conventions used in some variables of interest and where necessary (for consistency with reporting templates) modifies the names of these variables.
input_params_ls <- make_input_params(ds_tb, ds_descvs_ls = ds_descvs_ls, dv_ds_nm_and_url_chr = c("fakes", # Dataverse containing dataset "https://doi.org/10.7910/DVN/D74QMP"), # DOI header_yaml_args_ls = header_yaml_args_ls, maui_params_ls = maui_params_ls, output_format_ls = output_format_ls, predictors_lup = predictors_lup, prefd_covars_chr = "SOFAS", prefd_mdl_types_chr = c("OLS_CLL","GLM_GSN_LOG"), scndry_anlys_params_ls = scndry_anlys_params_ls)
Analyse, Report and Share
Run analysis
All analytic steps for our primary and secondary analyses are executed with the following function call. Note, this step can involve long execution times (potentially over an hour, depending on the input parameters).
write_analyses(input_params_ls)
Report results
We now generate the catalogues (one for the primary and one for each secondary analysis) that report the models created in the preceding step. The write_mdl_smry_rprt function also creates shareable versions (i.e. with copies of source dataset replaced by synthetic data) of the models. This step can take several minutes to execute.
input_params_ls <- write_mdl_smry_rprt(input_params_ls, use_shareable_mdls_1L_lgl = T)
Share results
We can now share the results created in the previous step by posting them to the online repository, the details of which we supplied earlier.
write_study_outp_ds(input_params_ls)
The write_study_outp_ds function writes its outputs to a draft dataset - which you can review by logging into your dataverse account, before clicking on the option to publish the dataset. The stucture of your draft dataset should be similar to those produced by this vignette, which are viewable at: https://doi.org/10.7910/DVN/D74QMP
Create manuscript
Create study metadata
The first step to creating a manuscript that provides the scientific summary of the study is to add some metadata about the study itself to our parameters list, using the make_study_descs_ls function. If you want the manuscript to refer to some of the candidate predictors by different names than that provided in the short_name_chr column of the predictors_lup object defined earlier, you can create a table of name correspondences and pass it to the var_nm_change_lup argument.
input_params_ls <- make_study_descs_ls(input_params_ls = input_params_ls, background_1L_chr = "Our study is entirely fictional and has been created to illustrate TTU package functionality.", coi_1L_chr = "None declared.", conclusion_1L_chr = "If this study was real, the results would be interesting.", ethics_1L_chr = "The study was reviewed and granted approval by Awesome University's Human Research Ethics Committee (1111111.1).", funding_1L_chr = "The study was funded by Generous Benefactor.", sample_desc_1L_chr = "The study sample is fake data that pretends to be young people aged 12 to 25 years who attended Australian primary care services for mental health related needs between November 2019 to August 2020.", time_btwn_bl_and_fup_1L_chr = "three months", var_nm_change_lup = tibble::tibble(old_nms_chr = c("PHQ9","GAD7"), new_nms_chr = c("PHQ-9", "GAD-7")))
Render auto-generated first-draft
We can now pass the updated input parameters list to the write_manuscript function. This will produce an initial draft of a scientific manuscript (detailed methods and results, a basic introduction and no discussion), by first downloading an extension to the TTU package (https://github.com/ready4-dev/ttu_lng_ss). If (AND ONLY IF), you are sure that you will never develop the manuscript further with a view to future publication, you can add this automatically generated draft to the study dataset by adding write_to_dv_1L_lgl = T to the function call (as below).
results_ls <- ready4show::write_manuscript(input_params_ls = input_params_ls, write_to_dv_1L_lgl = T)
The manuscript exported to the dataset by the preceding call is accessible at: https://dataverse.harvard.edu/api/access/datafile/4957407 . A local version will also automatically be saved on your machine and you can discover where this is by printing the object results_ls$path_params_ls$paths_ls$reports_dir_1L_chr.
You can override the output format specified in the input parameter list object by using the output_type_1L_chr argument. Note, that this time around we are using the results_ls object as the main input as this object was automatically saved (to the directory path stored in results_ls$path_params_ls$paths_ls$output_data_dir_1L_chr) by the preceding call.
Modify for publication
If you wish to develop the manuscript into something you will submit for publication, consult https://github.com/ready4-dev/ttu_lng_ss for guidance.
Purge dataset copies
As a final step, it is important to delete all local copies (but not the original version) of the study dataset that were produced during the analysis.
write_to_delete_ds_copies(results_ls)
\elandscape
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.