Extending the coverage of spectral libraries: a neighbor-based approach to predicting intensities of peptide fragmentation spectra

MS2CNN: predicting MS/MS spectrum based on protein sequence using deep convolutional neural networks

Background

Tandem mass spectrometry allows biologists to identify and quantify protein samples in the form of digested peptide sequences. When performing peptide identification, spectral library search is more sensitive than traditional database search but is limited to peptides that have been previously identified. An accurate tandem mass spectrum prediction tool is thus crucial in expanding the peptide space and increasing the coverage of spectral library search.

Results

We propose MS²CNN, a non-linear regression model based on deep convolutional neural networks, a deep learning algorithm. The features for our model are amino acid composition, predicted secondary structure, and physical-chemical features such as isoelectric point, aromaticity, helicity, hydrophobicity, and basicity. MS²CNN was trained with five-fold cross validation on a three-way data split on the large-scale human HCD MS² dataset of Orbitrap LC-MS/MS downloaded from the National Institute of Standards and Technology. It was then evaluated on a publicly available independent test dataset of human HeLa cell lysate from LC-MS experiments. On average, our model shows better cosine similarity and Pearson correlation coefficient (0.690 and 0.632) than MS²PIP (0.647 and 0.601) and is comparable with pDeep (0.692 and 0.642). Notably, for the more complex MS² spectra of 3+ peptides, MS²PIP is significantly better than both MS²PIP and pDeep.

Conclusions

We showed that MS²CNN outperforms MS²PIP for 2+ and 3+ peptides and pDeep for 3+ peptides. This implies that MS²CNN, the proposed convolutional neural network model, generates highly accurate MS² spectra for LC-MS/MS experiments using Orbitrap machines, which can be of great help in protein and peptide identifications. The results suggest that incorporating more data for deep learning model may improve performance.

Tandem mass spectral libraries of peptides and their roles in proteomics research

Proteomics is a rapidly maturing field aimed at the high-throughput identification and quantification of all proteins in a biological system. The cornerstone of proteomic technology is tandem mass spectrometry of peptides resulting from the digestion of protein mixtures. The fragmentation pattern of each peptide ion is captured in its tandem mass spectrum, which enables its identification and acts as a fingerprint for the peptide. Spectral libraries are simply searchable collections of these fingerprints, which have taken on an increasingly prominent role in proteomic data analysis. This review describes the historical development of spectral libraries in proteomics, details the computational procedures behind library building and searching, surveys the current applications of spectral libraries, and discusses the outstanding challenges. © 2016 Wiley Periodicals, Inc. Mass Spec Rev 36:634-648, 2017.

Impact of Amidination on Peptide Fragmentation and Identification in Shotgun Proteomics

Peptide amidination labeling using S-methyl thioacetimidate (SMTA) is investigated in an attempt to increase the number and types of peptides that can be detected in a bottom-up proteomics experiment. This derivatization method affects the basicity of lysine residues and is shown here to significantly impact the idiosyncracies of peptide fragmentation and peptide detectability. The unique and highly reproducible fragmentation properties of SMTA-labeled peptides, such as the strong propensity for forming b₁ fragment ions, can be further exploited to modify the scoring of peptide-spectrum pairs and improve peptide identification. To this end, we have developed a supervised postprocessing algorithm to exploit these characteristics of peptides labeled by SMTA. Our experiments show that although the overall number of identifications are similar, the SMTA modification enabled the detection of 16-26% peptides not previously observed in comparable CID/HCD tandem mass spectrometry experiments without SMTA labeling.

Quantitative Comparison of Tandem Mass Spectra Obtained on Various Instruments

The similarity between two tandem mass spectra, which were measured on different instruments, was compared quantitatively using the similarity index (SI), defined as the dot product of the square root of peak intensities in the respective spectra. This function was found to be useful for comparing energy-dependent tandem mass spectra obtained on various instruments. Spectral comparisons show the similarity index in a 2D "heat map", indicating which collision energy combinations result in similar spectra, and how good this agreement is. The results and methodology can be used in the pharma industry to design experiments and equipment well suited for good reproducibility. We suggest that to get good long-term reproducibility, it is best to adjust the collision energy to yield a spectrum very similar to a reference spectrum. It is likely to yield better results than using the same tuning file, which, for example, does not take into account that contamination of the ion source due to extended use may influence instrument tuning. The methodology may be used to characterize energy dependence on various instrument types, to optimize instrumentation, and to study the influence or correlation between various experimental parameters. Graphical Abstract ᅟ.

Spectral library searching in proteomics

Spectral library searching has become a mature method to identify tandem mass spectra in proteomics data analysis. This review provides a comprehensive overview of available spectral library search engines and highlights their distinct features. Additionally, resources providing spectral libraries are summarized and tools presented that extend experimental spectral libraries by simulating spectra. Finally, spectrum clustering algorithms are discussed that utilize the same spectrum-to-spectrum matching algorithms as spectral library search engines and allow novel methods to analyse proteomics data.

Machine learning applications in proteomics research: how the past can boost the future

Machine learning is a subdiscipline within artificial intelligence that focuses on algorithms that allow computers to learn solving a (complex) problem from existing data. This ability can be used to generate a solution to a particularly intractable problem, given that enough data are available to train and subsequently evaluate an algorithm on. Since MS-based proteomics has no shortage of complex problems, and since publicly available data are becoming available in ever growing amounts, machine learning is fast becoming a very popular tool in the field. We here therefore present an overview of the different applications of machine learning in proteomics that together cover nearly the entire wet- and dry-lab workflow, and that address key bottlenecks in experiment planning and design, as well as in data processing and analysis.