OffsampleAI: artificial intelligence approach to recognize off-sample mass spectrometry images.

BACKGROUND: Imaging mass spectrometry (imaging MS) is an enabling technology for spatial metabolomics of tissue sections with rapidly growing areas of applications in biology and medicine. However, imaging MS data is polluted with off-sample ions caused by sample preparation, particularly by the MALDI (matrix-assisted laser desorption/ionization) matrix application. Off-sample ion images confound and hinder statistical analysis, metabolite identification and downstream analysis with no automated solutions available. RESULTS: We developed an artificial intelligence approach to recognize off-sample ion images. First, we created a high-quality gold standard of 23,238 expert-tagged ion images from 87 public datasets from the METASPACE knowledge base. Next, we developed several machine and deep learning methods for recognizing off-sample ion images. The following methods were able to reproduce expert judgements with a high agreement: residual deep learning (F1-score 0.97), semi-automated spatio-molecular biclustering (F1-score 0.96), and molecular co-localization (F1-score 0.90). In a test-case study, we investigated off-sample images corresponding to the most common MALDI matrix (2,5-dihydroxybenzoic acid, DHB) and characterized properties of matrix clusters. CONCLUSIONS: Overall, our work illustrates how artificial intelligence approaches enabled by open-access data, web technologies, and machine and deep learning open novel avenues to address long-standing challenges in imaging MS.

Response Surface Methodology As a New Approach for Finding Optimal MALDI Matrix Spraying Parameters for Mass Spectrometry Imaging.

Automated spraying devices have become ubiquitous in laboratories employing matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI), in part because they permit control of a number of matrix application parameters that can easily be reproduced for intra- and interlaboratory studies. Determining the optimal parameters for MALDI matrix application, such as temperature, flow rate, spraying velocity, number of spraying cycles, and solvent composition for matrix application, is critical for obtaining high-quality MALDI-MSI data. However, there are no established approaches for optimizing these multiple parameters simultaneously. Instead optimization is performed iteratively (i.e., one parameter at a time), which is time-consuming and can lead to overall nonoptimal settings. In this report, we demonstrate the use a novel experimental design and the response surface methodology to optimize five parameters of MALDI matrix application using a robotic sprayer. Thirty-two combinations of MALDI matrix spraying conditions were tested, which allowed us to elucidate relationships between each of the application parameters as determined by MALDI-MS (specifically, using a 15 T Fourier transform ion cyclotron resonance mass spectrometer). As such, we were able to determine the optimal automated spraying parameters that minimized signal delocalization and enabled high MALDI sensitivity. We envision this optimization strategy can be utilized for other matrix application approaches and MALDI-MSI analyses of other molecular classes and tissue types.