diff --git a/DESCRIPTION b/DESCRIPTION
index af973548e..fbbfc59c3 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -14,7 +14,6 @@ License: Artistic-2.0
 Encoding: UTF-8
 LazyData: true
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.3.3
 biocViews: MassSpectrometry, Proteomics, Software, DataImport, QualityControl
 Depends:
     R (>= 4.0)
@@ -82,3 +81,4 @@ Collate:
     'utils_logging.R'
     'utils_shared_peptides.R'
 VignetteBuilder: knitr
+Config/roxygen2/version: 8.0.0
diff --git a/R/utils_documentation.R b/R/utils_documentation.R
index 3bf208f71..59b64b3d7 100644
--- a/R/utils_documentation.R
+++ b/R/utils_documentation.R
@@ -5,7 +5,11 @@
 #' @param removeFewMeasurements TRUE (default) will remove the features that have 1 or 2 measurements across runs.
 #' @param useUniquePeptide TRUE (default) removes peptides that are assigned for more than one proteins. 
 #' We assume to use unique peptide for each protein.
-#' @param summaryforMultipleRows max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.
+#' @param summaryforMultipleRows max or sum - when multiple PSMs identify
+#' the same feature within a single MS run (duplicate PSMs), use the
+#' highest (max) or sum of the duplicate intensities. Default is max for
+#' label-free converters and sum for TMT converters. Note that this parameter 
+#' does NOT control collapsing across fractions of the same biological mixture.
 #' @param removeProtein_with1Feature TRUE will remove the proteins which have only 1 feature, which is the combination of peptide, precursor charge, fragment and charge. FALSE is default.
 #' @param removeProtein_with1Peptide TRUE will remove the proteins which have only 1 peptide and charge. FALSE is default.
 #' @param removeOxidationMpeptides TRUE will remove the peptides including 'oxidation (M)' in modification. FALSE is default.
diff --git a/man/DIAUmpiretoMSstatsFormat.Rd b/man/DIAUmpiretoMSstatsFormat.Rd
index 60ddb5623..19f8d2339 100644
--- a/man/DIAUmpiretoMSstatsFormat.Rd
+++ b/man/DIAUmpiretoMSstatsFormat.Rd
@@ -38,7 +38,11 @@ DIAUmpiretoMSstatsFormat(
 
 \item{removeProtein_with1Feature}{TRUE will remove the proteins which have only 1 feature, which is the combination of peptide, precursor charge, fragment and charge. FALSE is default.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/FragPipetoMSstatsFormat.Rd b/man/FragPipetoMSstatsFormat.Rd
index f2b2e86f4..3553e61c7 100644
--- a/man/FragPipetoMSstatsFormat.Rd
+++ b/man/FragPipetoMSstatsFormat.Rd
@@ -27,7 +27,11 @@ We assume to use unique peptide for each protein.}
 
 \item{removeProtein_with1Feature}{TRUE will remove the proteins which have only 1 feature, which is the combination of peptide, precursor charge, fragment and charge. FALSE is default.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/MSstatsConvert.Rd b/man/MSstatsConvert.Rd
index e87142964..1a2e9ca25 100644
--- a/man/MSstatsConvert.Rd
+++ b/man/MSstatsConvert.Rd
@@ -23,6 +23,7 @@ signal processing tools to a format suitable for statistical analysis with the M
 
 Authors:
 \itemize{
+  \item Anthony Wu \email{wu.anthon@northeastern.edu}
   \item Mateusz Staniak \email{mtst@mstaniak.pl}
   \item Devon Kohler \email{kohler.d@northeastern.edu}
   \item Meena Choi \email{mnchoi67@gmail.com}
diff --git a/man/MaxQtoMSstatsFormat.Rd b/man/MaxQtoMSstatsFormat.Rd
index dd22234e9..3e154cbac 100644
--- a/man/MaxQtoMSstatsFormat.Rd
+++ b/man/MaxQtoMSstatsFormat.Rd
@@ -34,7 +34,11 @@ MaxQtoMSstatsFormat(
 \item{useUniquePeptide}{TRUE (default) removes peptides that are assigned for more than one proteins.
 We assume to use unique peptide for each protein.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{removeFewMeasurements}{TRUE (default) will remove the features that have 1 or 2 measurements across runs.}
 
diff --git a/man/MaxQtoMSstatsTMTFormat.Rd b/man/MaxQtoMSstatsTMTFormat.Rd
index af761b6be..aa468e64e 100644
--- a/man/MaxQtoMSstatsTMTFormat.Rd
+++ b/man/MaxQtoMSstatsTMTFormat.Rd
@@ -39,7 +39,11 @@ We assume to use unique peptide for each protein.}
 
 \item{rmProtein_with1Feature}{TRUE will remove the proteins which have only 1 peptide and charge. Default is FALSE.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/MetamorpheusToMSstatsFormat.Rd b/man/MetamorpheusToMSstatsFormat.Rd
index 1e8851be7..55333e30f 100644
--- a/man/MetamorpheusToMSstatsFormat.Rd
+++ b/man/MetamorpheusToMSstatsFormat.Rd
@@ -36,7 +36,11 @@ We assume to use unique peptide for each protein.}
 
 \item{removeProtein_with1Feature}{TRUE will remove the proteins which have only 1 feature, which is the combination of peptide, precursor charge, fragment and charge. FALSE is default.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/OpenMStoMSstatsFormat.Rd b/man/OpenMStoMSstatsFormat.Rd
index 2f8c39710..f72239444 100644
--- a/man/OpenMStoMSstatsFormat.Rd
+++ b/man/OpenMStoMSstatsFormat.Rd
@@ -31,7 +31,11 @@ We assume to use unique peptide for each protein.}
 
 \item{removeProtein_with1Feature}{TRUE will remove the proteins which have only 1 feature, which is the combination of peptide, precursor charge, fragment and charge. FALSE is default.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/OpenSWATHtoMSstatsFormat.Rd b/man/OpenSWATHtoMSstatsFormat.Rd
index 74ed2a3e2..a37e507a9 100644
--- a/man/OpenSWATHtoMSstatsFormat.Rd
+++ b/man/OpenSWATHtoMSstatsFormat.Rd
@@ -37,7 +37,11 @@ We assume to use unique peptide for each protein.}
 
 \item{removeProtein_with1Feature}{TRUE will remove the proteins which have only 1 feature, which is the combination of peptide, precursor charge, fragment and charge. FALSE is default.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/PDtoMSstatsFormat.Rd b/man/PDtoMSstatsFormat.Rd
index 40dbe8c05..d220d9ecf 100644
--- a/man/PDtoMSstatsFormat.Rd
+++ b/man/PDtoMSstatsFormat.Rd
@@ -34,7 +34,11 @@ Run information. 'Run' will be matched with 'Spectrum.File'.}
 \item{useUniquePeptide}{TRUE (default) removes peptides that are assigned for more than one proteins.
 We assume to use unique peptide for each protein.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{removeFewMeasurements}{TRUE (default) will remove the features that have 1 or 2 measurements across runs.}
 
diff --git a/man/PDtoMSstatsTMTFormat.Rd b/man/PDtoMSstatsTMTFormat.Rd
index 46a4cc561..e26438085 100644
--- a/man/PDtoMSstatsTMTFormat.Rd
+++ b/man/PDtoMSstatsTMTFormat.Rd
@@ -37,7 +37,11 @@ We assume to use unique peptide for each protein.}
 
 \item{rmProtein_with1Feature}{TRUE will remove the proteins which have only 1 peptide and charge. Default is FALSE.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/PhilosophertoMSstatsTMTFormat.Rd b/man/PhilosophertoMSstatsTMTFormat.Rd
index d2c8225a7..fa2ae66a7 100644
--- a/man/PhilosophertoMSstatsTMTFormat.Rd
+++ b/man/PhilosophertoMSstatsTMTFormat.Rd
@@ -50,7 +50,11 @@ We assume to use unique peptide for each protein.}
 
 \item{rmProtein_with1Feature}{TRUE will remove the proteins which have only 1 peptide and charge. Default is FALSE.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/ProgenesistoMSstatsFormat.Rd b/man/ProgenesistoMSstatsFormat.Rd
index 3d4cc7caf..aa3a8307f 100644
--- a/man/ProgenesistoMSstatsFormat.Rd
+++ b/man/ProgenesistoMSstatsFormat.Rd
@@ -27,7 +27,11 @@ ProgenesistoMSstatsFormat(
 \item{useUniquePeptide}{TRUE (default) removes peptides that are assigned for more than one proteins.
 We assume to use unique peptide for each protein.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{removeFewMeasurements}{TRUE (default) will remove the features that have 1 or 2 measurements across runs.}
 
diff --git a/man/ProteinProspectortoMSstatsTMTFormat.Rd b/man/ProteinProspectortoMSstatsTMTFormat.Rd
index 9e9d49db2..e2032fcac 100644
--- a/man/ProteinProspectortoMSstatsTMTFormat.Rd
+++ b/man/ProteinProspectortoMSstatsTMTFormat.Rd
@@ -32,7 +32,11 @@ We assume to use unique peptide for each protein.}
 
 \item{removeProtein_with1Feature}{TRUE will remove the proteins which have only 1 feature, which is the combination of peptide, precursor charge, fragment and charge. FALSE is default.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/SpectroMinetoMSstatsTMTFormat.Rd b/man/SpectroMinetoMSstatsTMTFormat.Rd
index cca8cb167..efbf4bec3 100644
--- a/man/SpectroMinetoMSstatsTMTFormat.Rd
+++ b/man/SpectroMinetoMSstatsTMTFormat.Rd
@@ -36,7 +36,11 @@ We assume to use unique peptide for each protein.}
 
 \item{rmProtein_with1Feature}{TRUE will remove the proteins which have only 1 peptide and charge. Defaut is FALSE.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{use_log_file}{logical. If TRUE, information about data processing
 will be saved to a file.}
diff --git a/man/SpectronauttoMSstatsFormat.Rd b/man/SpectronauttoMSstatsFormat.Rd
index dea2e4b99..91caca283 100644
--- a/man/SpectronauttoMSstatsFormat.Rd
+++ b/man/SpectronauttoMSstatsFormat.Rd
@@ -75,7 +75,11 @@ We assume to use unique peptide for each protein.}
 
 \item{removeProtein_with1Feature}{TRUE will remove the proteins which have only 1 feature, which is the combination of peptide, precursor charge, fragment and charge. FALSE is default.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{calculateAnomalyScores}{Default is FALSE. If TRUE, will run anomaly detection model and calculate anomaly scores for each feature. Used downstream to weigh measurements in differential analysis.}
 
diff --git a/man/dot-getFullDesign.Rd b/man/dot-getFullDesign.Rd
index a9f4ecbfc..f7b2720c3 100644
--- a/man/dot-getFullDesign.Rd
+++ b/man/dot-getFullDesign.Rd
@@ -13,12 +13,12 @@
 \code{feature_col} and \code{measurement_col} will be created within each unique value
 of this column}
 
-\item{is_tmt}{if TRUE, data will be treated as coming from TMT experiment.}
-
-\item{`feature_col`}{name of the column that labels features}
+\item{feature_col}{name of the column that labels features}
 
-\item{`measurement_col`}{name of a column with measurement labels - Runs in
+\item{measurement_col}{name of a column with measurement labels - Runs in
 label-free case, Channels in TMT case.}
+
+\item{is_tmt}{if TRUE, data will be treated as coming from TMT experiment.}
 }
 \value{
 data.table
diff --git a/man/dot-sharedParametersAmongConverters.Rd b/man/dot-sharedParametersAmongConverters.Rd
index c9e7c6cd7..9bf0ab9cc 100644
--- a/man/dot-sharedParametersAmongConverters.Rd
+++ b/man/dot-sharedParametersAmongConverters.Rd
@@ -12,7 +12,11 @@
 \item{useUniquePeptide}{TRUE (default) removes peptides that are assigned for more than one proteins.
 We assume to use unique peptide for each protein.}
 
-\item{summaryforMultipleRows}{max or sum - when there are multiple measurements for certain feature and certain run, use highest or sum of multiple intensities. Default is max for label-free converters and sum for TMT converters.}
+\item{summaryforMultipleRows}{max or sum - when multiple PSMs identify
+the same feature within a single MS run (duplicate PSMs), use the
+highest (max) or sum of the duplicate intensities. Default is max for
+label-free converters and sum for TMT converters. Note that this parameter
+does NOT control collapsing across fractions of the same biological mixture.}
 
 \item{removeProtein_with1Feature}{TRUE will remove the proteins which have only 1 feature, which is the combination of peptide, precursor charge, fragment and charge. FALSE is default.}
 
diff --git a/vignettes/msstats_data_format.Rmd b/vignettes/msstats_data_format.Rmd
index dbf77916a..17bec0d53 100644
--- a/vignettes/msstats_data_format.Rmd
+++ b/vignettes/msstats_data_format.Rmd
@@ -304,13 +304,29 @@ should consists of elements named
 
 # Fractions and balanced design
 
-Finally, after preprocessing, `MSstatsBalancedDesign` function can be applied to 
-handle fractions and create balanced design.
-For label-free and SRM data, it means that fractionation or technical replicates will be detected if these information is not provided. Features measured in multiple fractions (overlapped) will be assigned to a unique fraction. Then, the data will be adjusted so that within each fraction, every feature has a row for certain run. If the intensity value is missing, it will be denoted by `NA`.
-
-For TMT data, a unique fraction will be selected for each overlapped feature and the 
-data will adjusted so that within each run, every feature has a row for each channel.
-If the intensity is missing for a channel, it will be denoted by `NA`.
+Finally, after preprocessing, the `MSstatsBalancedDesign` function can be
+applied to handle fractions and create a balanced design.
+
+For label-free data, fractionation or technical replicates are
+detected if this information is not provided. Features that overlap
+across multiple fractions of the same sample are assigned to a single
+fraction by the following rule: for each feature, the fraction with the
+**largest number of MS runs containing a non-missing measurement** is
+kept. If multiple fractions tie on that count, the tie is broken by
+choosing the fraction with the **highest mean intensity**. The
+remaining fractions' rows for that feature are dropped. The data are
+then adjusted so that within each fraction, every feature has a row
+for each run. If the intensity value is missing, it is denoted by `NA`.
+
+For TMT data, a unique fraction is selected for each overlapped feature
+as well. After fraction selection, the data are adjusted so
+that within each run, every feature has a row for each channel. If
+the intensity is missing for a channel, it is denoted by `NA`.
+
+Note also that this fraction-collapsing logic is distinct from the
+`summaryforMultipleRows` argument on each converter, which only
+combines duplicate PSMs identifying the same feature within a single
+MS run.
 
 ```{r }
 maxquant_balanced = MSstatsBalancedDesign(maxquant_processed, feature_columns)