DDA-BERT Source Data

Description

This dataset provides the source data for the manuscript related to DDA-BERT. It contains the underlying data for all figures and tables presented in the main text and supplementary information.

File Organization

1. Source_data.xlsx

   • This primary Excel file contains the majority of the raw and processed data used to generate the charts and figures.

   • Data is organized into individual worksheets, each labeled according to its corresponding Figure or Supplementary Figure number.

2. Raw Data Archives (ZIP files) For figures involving large-scale datasets or numerous individual files (e.g., multi-species benchmark metrics), source data are archived into separate ZIP packages:

   • Figure2.zip: Contains scoring data for human samples across different software tools:

     • Figure2A-B_FragPipe_human_score.csv

     • Figure2A-B_Sage_human_score.csv

     • Figure2A-C_DDA-BERT_human_score.csv

   • Figure6.zip: Contains source files for diagnostic analysis of overfitting and sequence memorization in DDA-BERT, along with four benchmark files used for FDP (False Discovery Proportion) calculation:

     • Figure6A_box_plot.csv

     • Figure6B_intensity.csv

     • Figure6C_220627_GG_02_rep3_dil10_peptide.csv

     • Figure6C_220627_GG_04_rep1_dil10_peptide.csv

     • Figure6C_220627_GG_06_rep1_dil10_peptide.csv

     • Figure6C_220627_GG_07_rep2_dil10_peptide.csv

   • Supplementary Figure1.zip: Provides multi-species scoring data (fruit fly, Arabidopsis, and yeast) used for comparative analysis of different software tools:

     • SFigure1A_DDA-BERT/FragPipe/Sage_fruit_fly_score.csv

     • SFigure1B_DDA-BERT/FragPipe/Sage_artha_score.csv

     • SFigure1C_DDA-BERT/FragPipe/Sage_yeast_score.csv

   • Supplementary Figure4.zip: Contains summary tables of protein groups and peptide identification counts across all benchmark datasets, systematically organized by software tool:

     • Sub-folders: AlphaPept, AlphaPeptDeep, DDA-BERT, FP (FragPipe), MS2Rescore, and Sage.

Code and Visualization Scripts

The scripts used for data processing, analysis, and figure generation are available on the official GitHub repository:

• Repository: zhiyuajun/DDA-BERT

• Path: DDA-BERT/scripts

Model and Software Information

For detailed implementation of the DDA-BERT