release 1.2.2

• remove_proteins_by_name() now has some more flexibility for filtering undesired proteins from input, e.g. proteins from known contaminants/crapome FASTA.

• differential_detection_filter() is a new function for finding/filtering proteins that are detected in N samples in one sample ground and with less than M samples in another sample group. It also includes a cutoff/threshold for the ratio of detected peptides (per protein, across all samples) between sample groups.

• importing FASTA files now replaces semicolon symbols in headers with colons because we use semicolons as delimiter symbols (only recently have semicolon symbols been appearing in uniprot FASTA files, e.g. >tr|B1AR09|B1AR09_MOUSE Isoform of B1AR10, Myeloid/lymphoid or mixed-lineage leukemia; translocated to, 6 OS=Mus musculus OX=10090 GN=Mllt6 PE=1 SV=1)

• bugfix for plotting Cscore histograms in Spectronaut datasets to deal with samples that contain very few observed peptides

• bugfix for "differential detection" to deal with datasets that contain any sample with 0 detected peptides

• bugfix for reducing the RAM and RData footprint of large Spectronaut datasets

release 1.2.1

new function merge_proteingroups_by_gene() that can be used to merge proteingroups that map to the same gene symbol (isoforms). To be used immediately after importing a dataset and the respective FASTA files, so all downstream analyses will be applied to the merged proteingroups.

After updating MS-DAP, use ?merge_proteingroups_by_gene to view the documentation.

release 1.2

• add_contrast() is a new function that introduces additional flexibility for defining statistical contrasts
  • allows you to add a contrast with a wildcard for the second group (e.g. one may define group1 = "WT" and group2 = NA to create a contrast "WT" versus all other values in the "group" column)
    • also enables creating statistical contrasts based on other columns in the sample metadata table (dataset$samples) besides the default "group" column
    • after installing the update, run library(msdap) and check documentation with R command ?add_contrast()
  • this is accompanied by helper functions remove_contrasts() and print_contrasts()
  • you can still use the setup_contrasts() function as per usual, i.e. pre-existing MS-DAP scripts still work as intended
• new workflow for creating limma analyses (eBayes/DEqMS) with any custom model matrix / linear regression design
  • enables modeling of more complex experimental designs, e.g. paired samples / block designs or multi-level experiments
  • documentation and examples are available at bioinformatics: advanced statistical models with limma

release 1.1.3

• DIA-NN reports stored in parquet format can now be loaded with import_dataset_diann() (issue #34)
• gene symbols extracted from FASTA files with import_fasta() can now optionally be kept in original case by setting uppercase_symbols = FALSE (issue #35)

release 1.1.2

minor bug fixes related to FASTA file parsing (relates to regression bugs introduced in version 1.1)

release 1.1.1

• updated FragPipe import function; now uses quantitative data from the MSstats.csv file instead of collecting data across all ion.tsv files. This should ensure we use the intended output/filtered quantitative data and make our import function more robust (backtested on all major FragPipe releases from 14.0 to 22.0). Specifically, we import precursor ion intensities from MSstats.csv for all precursor*sample pairs, then search for respective PSM confidence scores and retention times in the psm.tsv files (not available for MBR hits)
• bugfix for parsing FASTA files that contain decoy/reverse entries; decoys are now properly recognized and only target proteins are imported into MS-DAP

release 1.1

• functions for gene-level summarizing of statistical results have been reworked
  • an introduction and quickstart guide is available on the landing page (README.md) of this GitHub repository, section "Creating gene-level summaries of statistical analysis results"
  • detailed documentation is available in R (after updating the msdap package): ?export_stats_genesummary
• effectsize columns were added to output file differential_abundance_analysis.xlsx
• differential detection analyses: default z-score threshold is now set to 6

release 1.0.9

• minor updates to the leave-one-out summary plot, includes a bugfix for large datasets

release 1.0.8

• bugfix for plot_peptide_data()

release 1.0.7

Major:

• bugfix; ggplot 3.5.0 introduced a few breaking changes for MS-DAP, these have been fixed

Minor:

• "within-group foldchange distributions" figures now use a robust summary statistic (trimmed means instead of mean) when 5+ replicates are available in a sample group
• in DDA datasets, "differential detect" plots now have independent x- and y-axis scales in each panel
• in DIA datasets, "differential detect" plot dimensions have been updated (slightly lowered height to sync with DDA variant)
• various minor updates/fixes for edgecases (e.g. datasets with groups/conditions that contain just 1 replicate)