Biology subjects

Burdukiewicz, M.

Publications and source records attributed to Burdukiewicz, M..

6 recordsLinked to original sources

bioRxiv2026

Application of class-balancing algorithms to diverse plasma metabolomics datasets using brain tumor as an example

Class imbalance remains a challenge in metabolomics research, where biological and technical variability can affect statistical inference and machine learning (ML) performance. Class-balancing algorithms address this issue by either increasing minority-class observations or reducing the number of majority-class samples. This study evaluated the impact of oversampling and undersampling algorithms on targeted and untargeted metabolomics datasets derived from LC-MS and GC-MS analyses of plasma samples from patients with glioblastoma, meningioma, and controls. Synthetic Minority Oversampling Technique (SMOTE) and Random Undersampling (RUS) were applied to balance the datasets, and their effects on data distribution, inter-feature correlations, and machine learning model performance were compared. RUS preserved the original feature distributions but reduced representativeness by removing the majority-class samples. In contrast, SMOTE introduced synthetic samples that altered covariance structures, increasing the risk of overfitting, particularly in small datasets (n=10). These effects diminished with larger groups (n=30), partially restoring correlations between metabolites. Model performance varied across the class-balancing algorithms. Random Forest classifiers benefited from both balancing methods, with undersampling often yielding higher F1 scores, whereas Support Vector Machine models showed reduced classification performance. These findings highlight the importance of selecting class-balancing strategies based on dataset size, analytical platform, and ML algorithm in metabolomics studies.

bioinformatics↗

bioRxiv2024

Scientific reasoning driven by influential data: resuscitate dfstat!

In biomedical literature, one of the most widely employed statistical procedures to analyze and visualize the association between two variables is linear regression. Data points that exert influence on the fit and its parameters are routinely, but not as often as required, identified by established influence measures and their corresponding cut-off values. In this work, we are specifically concerned with the presence of influential data points that directly impact hypothesis testing of linear regressions, which none of the established measures describe. Interestingly, the highly overlooked influence measure dfstat and its derived leave-one-out p-value exists exactly for this purpose, unmentioned in the majority of statistical text books as well as absent from all available statistical software packages. Its application for identifying these data points seems pivotal, as scientific reasoning in publications is almost exclusively based on the p-value of the fit, commonly adhering to the = 0.05 threshold to state significance or not. With this metric, we found for 29 of 100 digitizable papers published in Science, Nature and PNAS in 2016, a time when the "reproducibility crisis" was a growing concern, that stated significances (or their absence) are based on the presence of a single influential data point.

scientific communication and education↗

bioRxiv2022

The impact of negative data sampling on antimicrobial peptide prediction

Antimicrobial peptides (AMPs) are a heterogeneous group of short polypeptides that target microorganisms but also viruses and cancer cells. Due to their lower selection for resistance compared to traditional antibiotics, AMPs have been attracting the ever-growing attention from researchers, including bioinformaticians. Machine learning represents the most cost-effective method for novel AMP discovery and consequently many computational tools for AMP prediction have been recently developed. In this article, we investigate the impact of negative data sampling on model performance and benchmarking. We generated 660 predictive models using 12 machine learning architectures, a single positive data set and 11 negative data sampling methods; the architectures and methods were defined on the basis of published AMP prediction software. Our results clearly indicate that similar training and benchmark data set, i.e. produced by the same or a similar negative data sampling method, positively affect model performance. Consequently, all the benchmark analyses that have been performed for AMP prediction models are significantly biased and, moreover, we do not know which model is the most accurate. To provide researchers with reliable information about the performance of AMP predictors, we also created a web server AMPBenchmark for fair model benchmarking. AMPBenchmark is available at http://BioGenies.info/AMPBenchmark.

bioinformatics↗

bioRxiv2021

Functional amyloids in the microbiomes of a rat Parkinson's disease model and wild-type rats

Cross-seeding between amyloidogenic proteins in the gut is receiving increasing attention as a possible mechanism for initiation or acceleration of amyloid formation by aggregation-prone proteins such as SN, which is central in the development of Parkinsons disease. This is particularly pertinent in view of the growing number of functional (i.e. benign and useful) amyloid proteins discovered in bacteria. Here we identify two functional amyloid proteins, Pr12 and Pr17, in fecal matter from Parkinsons disease transgenic rats and their wild type counterparts, based on their stability against dissolution by formic acid. Both proteins show robust aggregation into ThT-positive aggregates that contain higher-order {beta}-sheets and have a fibrillar morphology, indicative of amyloid proteins. In addition, Pr17 aggregates formed in vitro showed significant resistance against formic acid, suggesting an ability to form highly stable amyloid. Treatment with proteinase K revealed a protected core of approx. 9 kDa. Neither Pr12 nor Pr17, however, affected SN aggregation in vitro. Thus, amyloidogenicity does not per se lead to an ability to cross-seed fibrillation of SN. Our results support the use of proteomics and formic acid to identify amyloid protein in complex mixtures and indicates the existence of numerous functional amyloid proteins in microbiomes. IMPORTANCEThe bacterial microbiome in the gastrointestinal tract is increasingly seen as important for human health and disease. One area of particular interest is that of neurodegenerative diseases such as Parkinsons which involve pathological aggregation into amyloid of human proteins such as - synuclein (SN). Bacteria are known to form benign or functional amyloid, some of which may initiate unwanted aggregation of e.g. SN in the enteric nervous system through cross-seeding via contact with the microbiome. Here we show that the rat microbiome contains several proteins which form this type of amyloid aggregate both in vivo and in vitro. Although the two proteins we investigate in depth do not directly promote SN aggregation, our work shows that the microbiome potentially harbors a significant number of bacterial amyloid which could play a role in human physiology at various levels.

biophysics↗

bioRxiv2021

PCRedux: A Data Mining and Machine Learning Toolkit for qPCR Experiments

MotivationQuantitative Real-time PCR (qPCR) is a widely used -omics method for the precise quantification of nucleic acids, in which the result is associated with the presence/absence or quantity of a specific nucleic acid sequence. As the amount of qPCR data increases worldwide, the manual assessment of results becomes challenging and difficult to reproduce. To overcome this, some automatable characteristics of amplification curves have been described in the literature, often with an appropriate "rule of thumb". ResultsWe developed PCRedux to analyze and calculate 90 numerical qPCR amplification curve descriptors ( features") from large datasets of qPCR amplification curves that are aimed for interpretable machine learning and development of decision support systems. In a case study of a diverse dataset with 3181 positive, negative and ambiguous amplification curves, as assessed by three human raters, we demonstrate a sensitivity >99 % and specificity >97 % in detecting positive and negative amplification. PCRedux is unique as it goes beyond traditional qPCR analysis to capture curvature properties that improve the characterization and classification of amplification curves. The calculation of the features is reproducible and objective, since R is used as a controllable working environment. PCRedux is not a black box, but open source software following on the principle of mathematically interpretable features. These can be combined with user-defined labels for automatic multi-category classification and regression in machine learning. Availabilityhttps://cran.r-project.org/package=PCRedux. Web server: http://shtest.evrogen.net/PCRedux-app/. Documentation: https://PCRuniversum.github.io/PCRedux/.

bioinformatics↗

bioRxiv2020

MegaGO: a fast yet powerful approach to assess functional similarity across meta-omics data sets

The study of microbiomes has gained in importance over the past few years, and has led to the fields of metagenomics, metatranscriptomics and metaproteomics. While initially focused on the study of biodiversity within these communities the emphasis has increasingly shifted to the study of (changes in) the complete set of functions available in these communities. A key tool to study this functional complement of a microbiome is Gene Ontology (GO) term analysis. However, comparing large sets of GO terms is not an easy task due to the deeply branched nature of GO, which limits the utility of exact term matching. To solve this problem, we here present MegaGO, a user-friendly tool that relies on semantic similarity between GO terms to compute functional similarity between two data sets. MegaGO is highly performant: each set can contain thousands of GO terms, and results are calculated in a matter of seconds. MegaGO is available as a web application at https://megago.ugent.be and installable via pip as a standalone command line tool and reusable software library. All code is open source under the MIT license, and is available at https://github.com/MEGA-GO/.

bioinformatics↗