Protein Identification Improvement in Complex Samples Using Higher Frequency MS Acquisition and PEAKS Software

Protein identification in complex biological samples using the shotgun mode of LC-MS/MS is typically enhanced by employing longer LC columns and extended gradient times. However, improved identification rates can also be achieved by optimizing MS acquisition frequencies and employing advanced software, without increasing analysis time, thus maintaining the throughput of the method. To date, we found only one study in the literature examining the influence of MS acquisition frequency on protein identification, specifically using two ion trap mass spectrometer models. This study aims to address the gap by analyzing the impact of MS acquisition tuning of the QTOF instrument on the analysis of complex samples. Our findings indicate that increasing acquisition frequency generally improves protein identification, although the extent of improvement depends on the sample type. For CHO cell lysates, protein identifications increased by over 10%, while E. coli and albumin-depleted plasma samples demonstrated gains of 3.6% and 2.6%, respectively. Higher contributions to protein identification were also achieved with extended LC gradients, resulting in improvements of 21.6% for CHO, 18.2% for E. coli, and 10.3% for plasma. Moreover, enabling PEAKS’ deep learning feature significantly boosted identifications, with increases of 22.9% for CHO, 23.2% for E. coli, and 9.2% for plasma.