Biology subjects

Markuzon, N.

Publications and source records attributed to Markuzon, N..

4 recordsLinked to original sources

bioRxiv2024

Integrating Knowledge Graphs into Machine Learning Models for Survival Prediction and Biomarker Discovery in Patients with Non-Small-Cell Lung Cancer

Survival prediction is a critical aspect of clinical study design and biomarker discovery. It is a highly complex task, given the large number of "omics" and clinical features, as well as the high degrees of freedom that drive patient survival. Prior knowledge can play a critical role in uncovering the complexity of a disease and understanding the driving factors affecting a patients survival. We introduce a methodology for incorporating prior knowledge into machine learning-based models for prediction of patient survival through knowledge graphs, demonstrating the advantage of such an approach for patients with non-small-cell lung cancer. Using data from patients treated with immuno-oncologic therapies in the POPLAR (NCT01903993) and OAK (NCT02008227) clinical trials, we found that the use of knowledge graphs yielded significantly improved hazard ratios, including in the POPLAR cohort, for models based on biomarker tumor mutation burden compared with those based on knowledge graphs. Use of a model-defined mutational 10-gene signature led to significant overall survival differentiation for both trials. We provide parameterized code for incorporating knowledge graphs into survival analyses for use by the wider scientific community.

bioinformatics↗

bioRxiv2024

Quantifying the advantage of multimodal data fusion for survival prediction in cancer patients

The last decade has seen an unprecedented advance in technologies at the level of high-throughput molecular assays and image capturing and analysis, as well as clinical phenotyping and digitization of patient data. For decades, genotyping (identification of genomic alterations), the casual anchor in biological processes, has been an essential component in interrogating disease progression and a guiding step in clinical decision making. Indeed, survival rates in patients tested with next-generation sequencing have been found to be significantly higher in those who received a genome-guided therapy than in those who did not. Nevertheless, DNA is only a small part of the complex pathophysiology of cancer development and progression. To assess a more complete picture, researchers have been using data taken from multiple modalities, such as transcripts, proteins, metabolites, and epigenetic factors, that are routinely captured for many patients. Multimodal machine learning offers the potential to leverage information across different bioinformatics modalities to improve predictions of patient outcome. Identifying a multiomics data fusion strategy that clearly demonstrates an improved performance over unimodal approaches is challenging, primarily due to increased dimensionality and other factors, such as small sample sizes and the sparsity and heterogeneity of data. Here we present a flexible pipeline for systematically exploring and comparing multiple multimodal fusion strategies. Using multiple independent data sets from The Cancer Genome Atlas, we developed a late fusion strategy that consistently outperformed unimodal models, clearly demonstrating the advantage of a multimodal fusion model.

bioinformatics↗

bioRxiv2021

Bi-phasic dynamics of the mitochondrial DNA mutation m.3243A>G in blood: An unbiased, mutation level-dependent model implies positive selection in the germline.

The A-to-G point mutation at position 3243 in the human mitochondrial genome (m.3243A>G) is the most common pathogenic mtDNA variant responsible for disease in humans. It is widely accepted that m.3243A>G levels decrease in blood with age, and an age correction representing [~]2% annual decline is often applied to account for this change in mutation level. Here we report that recent data indicate the dynamics of m.3243A>G are more complex and depend on the mutation level in blood in a bi-phasic way. Consequently, the traditional 2% correction, which is adequate on average, creates opposite predictive biases at high and low mutation levels. Unbiased age correction is needed to circumvent these drawbacks of the standard model. We propose to eliminate both biases by using an approach where age correction depends on mutation level in a biphasic way to account for the dynamics of m.3243A>G in blood. The utility of this approach was further tested in estimating germline selection of m.3243A>G. The biphasic approach permitted us to uncover patterns consistent with the possibility of positive selection for m.3243A>G. Germline selection of m.3243A>G shows an arching profile by which selection is positive at intermediate mutant fractions and declines at high and low mutant fractions. We conclude that use of this biphasic approach will greatly improve the accuracy of modelling changes in mtDNA mutation frequencies in the germline and in somatic cells during aging.

genomics↗

bioRxiv2017

Mitochondrial pseudogenes suggest repeated inter-species hybridization in hominid evolution.

The hypothesis that the evolution of humans involved hybridization between diverged species has been actively debated in recent years. We present novel evidence in support of this hypothesis: the analysis of nuclear pseudogenes of mtDNA (\"NUMTs\"). NUMTs are considered \"mtDNA fossils\", as they preserve sequences of ancient mtDNA and thus carry unique information about ancestral populations. Our comparison of a NUMT sequence shared by humans, chimpanzees, and gorillas with their mtDNAs implies that, around the time of divergence between humans and chimpanzees, our evolutionary history involved the interbreeding of individuals whose mtDNA had diverged as much as ~4.5 Myr prior. This large divergence suggests a distant interspecies hybridization. Additionally, analysis of two other NUMTs suggests that such events occurred repeatedly. Our findings suggest a complex pattern of speciation in primate human ancestors and provide a potential explanation for the mosaic nature of fossil morphology found at the emergence of the hominin lineage.

evolutionary biology↗