Crosslink mass spectrometry dataset on soluble, SEC-fractionated high-molecular weight proteome of E. coli, crosslinked with BS3 or DSSO. This data was used to identify protein-protein interactions and to obtain information on protein (complex) topologies. In the end, the data was used for finding and evaluating a reliable approach for error control in crosslink mass spectrometry, particularly for PPI-studies. Cleared E. coli lysate was fractionated in the range of 3 MDa to 150 kDa (44 fractions). After taking aliquots for quantitative proteomics (JPST000843 ), the aliquots were split and the proteins in each fraction crosslinked (with BS3 or DSSO) and pooled again for each crosslinker set. Then, the pool was precipitated, derivatized and proteolysed in-solution. Digests were fractionated by SCX and hSAX (9x10 offline-fractionation matrix). Afterward, each peptide fraction was acquired via LC-MS on Q Exactive HF. Raw data was pre-processed (denoising, peak-list generation, error correction) and this data searched with xiSEARCH 1.6.746 using differing databases. FDR was calculated using xiFDR (version 2.0dev) under various settings to probe several FDR approaches.