An antimicrobial drug recommender system using MALDI-TOF MS and dual-branch neural networks

De Waele, Gaetan; Menschaert, Gerben; Waegeman, Willem

doi:10.1101/2023.09.28.559916

Cited by 2 publications

(7 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DRIAMS contains a total of 250 070 spectra, originating from four hospitals in Switzerland. For AMR prediction, the same data splits are used as in earlier work (De Waele et al, 2023). Briefly, DRIAMS-A spectra from before 2018 are split to the training fraction, whereas DRIAMS-A spectra measured during 2018 are evenly split between validation and test set.…”

Section: Methodsmentioning

confidence: 99%

“…We train Maldi Transformer in four different sizes: S, M, L, and XL. Model sizes are chosen so that the total weight numbers roughly correspond to the ones in De Waele et al (2023). Table 2 lists the size of all models, along with some hyperparameter settings.…”

Section: Methodsmentioning

confidence: 99%

“…For AMR prediction, Maldi Transformer is used as a spectrum embedder in a dual-branch neural network recommender system. To compare its performance against previous results, recommenders are trained with the four best-scoring drug embedders in De Waele et al (2023). To benchmark Maldi Transformer in terms of species identification, the RKI and LM-UGent datasets are used.…”

Section: Methodsmentioning

confidence: 99%

“…For AMR prediction, training of Maldi Transformerbased recommenders is performed identical to the MLP-based baselines in De Waele et al (2023). Briefly explained, for every combination of spectrum embedder (four sizes: S, M, L, and XL) and drug embedder (four types), four different learning rates ({1e-5, 5e-5, 1e-4, 5e-4}) are tested.…”

Section: Driams Species Processingmentioning

confidence: 99%

“…Because of this, the nascent field has not often progressed beyond off-the-shelf learning techniques (Weis et al, 2020a). Only a handful of examples exist of more advanced machine learning methods specif ically adapted to a MALDI-TOF-based task (Vervier et al, 2015; Weis et al, 2020b; Mortier et al, 2021; De Waele et al, 2023). As such, the design of machine learning algorithms specifically for MALDI-TOF mass spectra has not been studied yet in sufficient detail.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Pre-trained Maldi Transformers improve MALDI-TOF MS-based prediction

De Waele,

Menschaert,

Vandamme

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

For the last decade, matrix-assisted laser desportion/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has been the reference method for species identification in clinical microbiology. Hampered by a historical lack of open data, machine learning research towards models specifically adapted to MALDI-TOF MS remains in its infancy. Given the growing complexity of available datasets (such as large-scale antimicrobial resistance prediction), a need for models that (1) are specifically designed for MALDI-TOF MS data, and (2) have high representational capacity, presents itself. Here, we introduce Maldi Transformer, an adaptation of the state-of-the-art transformer architecture to the MALDI-TOF mass spectral domain. We propose the first self-supervised pre-training technique adapted to mass spectra. The technique is based on shuffling peaks across spectra, and pre-training the transformer as a peak discriminator. Extensive benchmarks confirm the efficacy of this novel design. The final result is a model exhibiting state-of-the-art (or competitive) performance on downstream prediction tasks. In addition, we show that Maldi Transformer's identification of noisy spectra may be leveraged towards higher predictive performance. All code supporting this study is distributed on PyPI and is packaged under: https://github.com/gdewael/maldi-nn

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Driams Species Processingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Pre-trained Maldi Transformers improve MALDI-TOF MS-based prediction

De Waele,

Menschaert,

Vandamme

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Overcoming Challenges of Reproducibility and Variability for theClostridioides difficiletypification

Antón,

Guerrero-López,

Sevilla-Salcedo

et al. 2024

Preprint

View full text Add to dashboard Cite

The implementation of Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry has had a profound impact on clinical microbiology, facilitating rapid bacterial identification through protein profile analysis. However, the application of this technique is limited by challenges related to the reproducibility and variability of spectra, particularly in distinguishing closely related bacterial strains, as exemplified by the typification of Clostridioides difficile ribotypes. This paper investigates the integration of Machine Learning algorithms to enhance the robustness and accuracy of MALDI-TOF spectra analysis. The aim is to compare traditional classifiers in order to gain insight into how spectral variability affects their performance in typification. Furthermore, specific data augmentation techniques for MALDI-TOF spectra are designed to enhance the classification of C. difficile ribotypes, to alleviate the inherent variability in MALDI-TOF spectra, and to address the issue of limited sample sizes. The results demonstrate that these methods can significantly enhance the accuracy of classification of C. difficile strains, thereby rendering MALDI-TOF a more reliable tool in clinical diagnostics.

show abstract

An antimicrobial drug recommender system using MALDI-TOF MS and dual-branch neural networks

Cited by 2 publications

References 59 publications

Pre-trained Maldi Transformers improve MALDI-TOF MS-based prediction

Pre-trained Maldi Transformers improve MALDI-TOF MS-based prediction

Overcoming Challenges of Reproducibility and Variability for theClostridioides difficiletypification

Contact Info

Product

Resources

About