Biology subjects

Karathanasis, N.

Publications and source records attributed to Karathanasis, N..

6 recordsLinked to original sources

bioRxiv2024

Machine Learning Models for Predicting Multiple Myeloma Staging and MGUS Progression Using Gene Expression Data

In this study, we developed and evaluated Machine Learning (ML) models aimed at predicting the stage of multiple myeloma (MM) and the progression of monoclonal gammopathy of undetermined significance (MGUS) to MM. Accurate staging of MM is critical for determining appropriate treatment strategies, and our models, employing algorithms such as ElasticNet, Random Forest, Boosting, and Support Vector Machines, demonstrated high efficacy in capturing the biological differences across disease stages. Among these, the ElasticNet model exhibited strong generalizability, achieving consistent multiclass AUC values across various datasets and data transformations. Predicting MGUS progression to MM presents a significant challenge due to the scarcity of MGUS cases that have progressed. We employed a two-pronged approach to address this: developing models using a limited dataset containing progressing MGUS patients and training models on combined MGUS and MM datasets. The models achieved AUC values slightly above 0.8, particularly with ElasticNet, Boosting and Support Vector Machines, indicating their potential in stratifying MGUS patients by progression risk. This study is original in integrating MM data with MGUS cases to enhance the predictive accuracy of MGUS progression, offering a novel methodology with potential clinical applications in patient monitoring and early intervention. Our feature selection and enrichment analyses further revealed that the identified genes are involved in key signaling pathways, including PI3K-Akt, MAPK, Wnt, and mTOR, all of which play crucial roles in MM pathogenesis. These findings align with established biological knowledge, suggest possible therapeutic targets and increase the explainability of our models.

bioinformatics↗

bioRxiv2024

Stage-specific gene ratios highlight genes and mechanisms related to presymptomatic and symptomatic Multiple Myeloma

Background/aimMultiple Myeloma is the second most common blood cancer, characterised by the accumulation of malignant plasma cells and the production of large amounts of a monoclonal immunoglobulin protein, in the bone marrow. The identification and progression/behaviour of molecular markers across stages remains a scientific challenge. This work aims to provide a holistic approach to the understanding of the disease progression, providing specific methodologies and candidate biomarkers, able to characterise and distinguish the disease state across stages. Materials and methodsTwo large bulk RNA datasets were used to collect and integrate stage-specific information at the gene level by means of: (a) differential expression analysis to obtain differential expressed genes (DEGs), (b) a recently introduced computational methodology able to detect monotonically expressed genes (MEGs), (c) a proposed computational methodology that uses pairs of MEGs at sample level, to classify and discriminate different stages of Multiple Myeloma. Additional numerical metrics were applied to rank the performance of these pairs across samples, facilitating the characterization and differentiation of disease stages. Validation was conducted using five additional external datasets, which were then utilised to enrich the final selection of top-rated genes identified from the two bulk RNA datasets under study. The final top-ranked genes were further used for pathway enrichment analysis in order to provide candidate pathways per stage. ResultsWe first show that MEGs provide better statistics than DEGs, both at gene and pathway level analysis. Secondly, we show that the proposed computational methodology by means of MEGs reveals short lists of high discriminative genes across stages, which in turn highlight significant groups of pathways. ConclusionWe integrated traditional analysis of DEGs with a recently introduced methodology for identifying MEGs, creating a novel computational approach capable of identifying highly discriminative genes and pathways that can serve as candidate markers for stage identification in a single sample. HighlightsO_LIA novel computational approach was used to identify Monotonically Expressed Genes (MEGs) whose expression was constantly increasing or decreasing. Genes such as RB1, CD27, TP53, and MCL1, previously highlighted in Multiple Myeloma, showed a consistent monotonic pattern, providing potential indicators for tracking the progression from the pre-malignant stages to active Multiple Myeloma. C_LIO_LIThe study calculated gene pair ratios using MEGs characterised by normal distribution and low dispersion. These ratios effectively distinguished healthy from disease samples, although the discrimination between disease stages (MGUS, SMM, MM) was less clear due to their overlapping molecular profiles. C_LIO_LIEnrichment analysis of significant gene pairs identified critical pathways affected during Multiple Myeloma progression, such as bone disease-related calcium pathways, glucocorticoid-regulated functions, and cardiac and neurological systems. These findings align with known clinical manifestations in MM patients, such as bone disease and amyloid cardiomyopathy. C_LIO_LIThe gene lists generated from our computational approach were validated against internal and external datasets, confirming their applicability. The methodology showed promise in identifying candidate genes for disease progression and could be applied to other diseases to uncover novel gene pairs not highlighted by traditional analyses. C_LI

bioinformatics↗

bioRxiv2023

Ranking of cell clusters in a single-cell RNA-sequencing analysis framework using prior knowledge

Prioritization or ranking of different cell types in a scRNA-Seq framework can be performed in a variety of ways, some of these include: i) obtaining an indication of the proportion of cell types between the different conditions under study, ii) counting the number of differentially expressed genes (DEGs) between cell types and conditions in the experiment or, iii) prioritizing cell types based on prior knowledge about the conditions under study (i.e., a specific disease). These methods have drawbacks and limitations thus novel methods for improving cell ranking are required. Here we present a novel methodology that exploits prior knowledge in combination with expert-user information to accentuate cell types from a scRNA-seq analysis that yield the most biologically meaningful results. Prior knowledge is incorporated in a standardized, structured manner, whereby a checklist is attained by querying MalaCards human disease database with a disease of interest. The checklist is comprised of pathways and drugs and, optionally, drug mode of actions (MOAs), associated with the disease. The user is prompted to "edit" this checklist by removing or adding terms (in the form of keywords) from the list of predefined terms. Our methodology has substantial advantages to more traditional cell ranking techniques and provides an informative complementary methodology that utilizes prior knowledge in a rapid and automated manner, that has previously not been attempted by other studies. The current methodology is also implemented as an R package entitled Single Cell Ranking Analysis Toolkit (scRANK) and is available for download and installation via GitHub (https://#hub.com/aoulas/scRANK) Author SummarySingle-cell RNA Sequencing (scRNA-Seq) provides an additive resolution down to the cellular level that was previously not available from traditional "Bulk" RNA sequencing experiments. However, it is often difficult to prioritize the specific cell-types which are primarily responsible for the cause of a disease. This work presents a novel methodology that exploits prior knowledge for a disease in combination with expert-user information to accentuate cell types from a scRNA-seq analysis that are most closely related to the molecular mechanism of a disease of interest. Prior knowledge is incorporated in a standardized, structured manner, whereby a checklist is attained by querying a human disease database called MalaCards. This checklist comes in the form of pathways and drugs and, optionally, drug mode of actions (MOAs), associated with the disease. The expert-user is prompted to "edit" the prior knowledge checklist by removing or adding terms from the list of predefined terms. This methodology has substantial advantages to more traditional cell prioritizing or ranking techniques and provides an informative complementary methodology that utilizes prior knowledge in a rapid and automated manner.

bioinformatics↗

bioRxiv2020

STATegra: Multi-omics data integration - A conceptual scheme with a bioinformatics pipeline

Technologies for profiling samples using different omics platforms have been at the forefront since the human genome project. Large-scale multi-omics data hold the promise of deciphering different regulatory layers. Yet, while there is a myriad of bioinformatics tools, each multi-omics analysis appears to start from scratch with an arbitrary decision over which tools to use and how to combine them. It is therefore an unmet need to conceptualize how to integrate such data and to implement and validate pipelines in different cases. We have designed a conceptual framework (STATegra), aiming it to be as generic as possible for multi-omics analysis, combining machine learning component analysis, non-parametric data combination and a multi-omics exploratory analysis in a step-wise manner. While in several studies we have previously combined those integrative tools, here we provide a systematic description of the STATegra framework and its validation using two TCGA case studies. For both, the Glioblastoma and the Skin Cutaneous Melanoma cases, we demonstrate an enhanced capacity to identify features in comparison to single-omics analysis. Such an integrative multi-omics analysis framework for the identification of features and components facilitates the discovery of new biology. Finally, we provide several options for applying the STATegra framework when parametric assumptions are fulfilled, and for the case when not all the samples are profiled for all omics. The STATegra framework is built using several tools, which are being integrated step-by-step as OpenSource in the STATegRa Bioconductor package https://bioconductor.org/packages/release/bioc/html/STATegra.html.

bioinformatics↗

bioRxiv2019

Machine Learning Approaches Identify Genes Containing Spatial Information from Single-Cell Transcriptomics Data.

MotivationWe participated in the DREAM Single Cell Transcriptomics Challenge. The challenges focus was two-fold; a) to identify the top 60, 40 and 20 genes that contain the most spatial information, and b) to reconstruct the 3-D arrangement of the D. melanogaster embryo using information from those genes.\n\nResultsWe developed two independent approaches, leveraging machine learning models from Lasso and Deep Neural Networks, that we successfully apply to high-dimensional single-cell sequencing data. Our methods allowed us to achieve top performance when compared to the ground truth. Among ~40 participating teams, the resulting solutions placed 10th, 6th, and 4th in the three DREAM sub-challenges #1, #2 and #3, respectively. Notably, for the Lasso approach we introduced a feature selection technique, Lasso-TopX, that allows a user to define a specific number of features they are interested in and the Neural Network approach utilizes weak supervision for linear regression to accommodate for uncertain or probabilistic training labels. Furthermore, we identified novel D. melanogaster genes that carry important positional information and were not previously suspected. Lastly, we show how the indirect use of the full datasets information can lead to data leakage and generate bias in overestimating the models performance.\n\nAvailabilityhttps://github.com/TJU-CMC-Org/SingleCell-DREAM/.\n\nContactNestoras.Karathanasis@jefferson.edu

bioinformatics↗

bioRxiv2019

Predicting cellular position in the Drosophila embryo from Single-Cell Transcriptomics data

Single-cell RNA-seq technologies are rapidly evolving but while very informative, in standard scRNAseq experiments the spatial organization of the cells in the tissue of origin is lost. Conversely, spatial RNA-seq technologies designed to keep the localization of the cells have limited throughput and gene coverage. Mapping scRNAseq to genes with spatial information increases coverage while providing spatial location. However, methods to perform such mapping have not yet been benchmarked. To bridge the gap, we organized the DREAM Single-Cell Transcriptomics challenge focused on the spatial reconstruction of cells from the Drosophila embryo from scRNAseq data, leveraging as gold standard genes with in situ hybridization data from the Berkeley Drosophila Transcription Network Project reference atlas. The 34 participating teams used diverse algorithms for gene selection and location prediction, while being able to correctly localize rare subpopulations of cells. Selection of predictor genes was essential for this task and such genes showed a relatively high expression entropy, high spatial clustering and the presence of prominent developmental genes such as gap and pair-ruled genes and tissue defining markers.

systems biology↗