Biology subjects

Blumenthal, D. B.

Publications and source records attributed to Blumenthal, D. B..

7 recordsLinked to original sources

bioRxiv2024

Spatial cell graph analysis reveals skin tissue organization characteristic for cutaneous T cell lymphoma

Cutaneous T cell lymphomas (CTCLs) are non-Hodgkin lymphomas caused by malignant T cells which migrate to the skin. The cancerous T cells lead to rash-like lesions which can be difficult to distinguish from inflammatory skin conditions like atopic dermatitis (AD) and psoriasis (PSO). To characterize CTCL in comparison to these differential diagnoses, we carried out multi-antigen imaging on 69 skin tissue samples (21 CTCL, 23 AD, 25 PSO). The resulting spatially resolved protein abundance maps were then analyzed via scoring functions to quantify heterogeneity of the individual cells neighborhoods within spatial graphs inferred from the cells positions in the tissue samples (available as a Python package at https://github.com/bionetslab/SHouT). Our analyses reveal several characteristic patterns of skin tissue organization in CTCL, including a combination of increased local entropy and egophily as characteristic properties of spatial T cell neighborhoods in CTCL as compared to AD and PSO. These results could not only pave the way for high-precision diagnosis of CTCL, but may also facilitate further insights into cellular disease mechanisms.

bioinformatics↗

bioRxiv2024

Identification of transcriptional regulators using a combined disease module identification and prize-collecting Steiner tree approach

Transcription factors play important roles in maintaining normal biological function, and their dys-regulation can lead to the development of diseases. Identifying candidate transcription factors involved in disease pathogenesis is thus an important task for deriving mechanistic insights from gene expression data. We developed Transcriptional Regulator Identification using Prize-collecting Steiner trees (TRIPS), a workflow for identifying candidate transcriptional regulators from case-control expression data. In the first step, TRIPS combines the results of differential expression analysis with a disease module identification step to retrieve perturbed subnetworks comprising an expanded gene list. TRIPS then solves a prize-collecting Steiner tree problem on a gene regulatory network, thereby identifying candidate transcriptional modules and transcription factors. We compare TRIPS to relevant methods using publicly available disease datasets and show that the proposed workflow can recover known disease-associated transcription factors with high precision. Network perturbation analyses demonstrate the reliability of TRIPS results. We further evaluate TRIPS on Alzheimers disease, diabetic kidney disease, and prostate cancer single-cell omics datasets. Overall, TRIPS is a useful approach for prioritizing transcriptional mechanisms for further downstream analyses.

bioinformatics↗

bioRxiv2023

DataSAIL: Data Splitting Against Information Leakage

Information Leakage is an increasing problem in machine learning research. It is a common practice to report models with benchmarks, comparing them to the state-of-the-art performance on the test splits of datasets. If two or more dataset splits contain identical or highly similar samples, a model risks simply memorizing them, and hence, the true performance is overestimated, which is one form of Information Leakage. Depending on the application of the model, the challenge is to find splits that minimize the similarity between data points in any two splits. Frequently, after reducing the similarity between training and test sets, one sees a considerable drop in performance, which is a signal of removed Information Leakage. Recent work has shown that Information Leakage is an emerging problem in model performance assessment. This work presents DataSAIL, a tool for splitting biological datasets while minimizing Information Leakage in different settings. This is done by splitting the dataset such that the total similarity of any two samples in different splits is minimized. To this end, we formulate data splitting as a Binary Linear Program (BLP) following the rules of Disciplined Quasi-Convex Programming (DQCP) and optimize a solution. DataSAIL can split one-dimensional data, e.g., for property prediction, and two-dimensional data, e.g., data organized as a matrix of binding affinities between two sets of molecules, accounting for similarities along each dimension and missing values. We compute splits of the MoleculeNet benchmarks using DeepChem, the LoHi splitter, GraphPart, and DataSAIL to compare their computational speed and quality. We show that DataSAIL can impose more complex learning tasks on machine learning models and allows for a better assessment of how well the model generalizes beyond the data presented during training.

bioinformatics↗

bioRxiv2023

Emergence of power-law distributions in protein-protein interaction networks through study bias

Protein-protein interaction (PPI) networks are power-law-distributed. However, the experimental procedures for detecting PPIs are affected by technical and study bias. For instance, cancer-associated proteins have received disproportional attention. Moreover, bait proteins in large-scale experiments tend to have many false-positive interaction partners. This raises the question of whether PL distributions in observed PPI networks could be explained by these biases alone. To assess this question, we studied the degree distribution of thousands of PPI networks of controlled provenance. Our findings are supported by mathematical models and extensive simulations and indicate that study bias and technical bias suffice to produce the observed PL distribution. It is, hence, problematic to derive hypotheses about the degree distribution and the true biological interactome from the PL distributions in observed PPI networks. Our study casts doubt on the use of the PL property of biological networks as a modeling assumption or quality criterion in network biology.

systems biology↗

bioRxiv2023

Cracking the black box of deep sequence-based protein-protein interaction prediction

Identifying protein-protein interactions (PPIs) is crucial for deciphering biological pathways. Numerous prediction methods have been developed as cheap alternatives to biological experiments, reporting surprisingly high accuracy estimates. We systematically investigated how much reproducible deep learning models depend on data leakage, sequence similarities, and node degree information, and compared them to basic machine learning models. We found that overlaps between training and test sets resulting from random splitting lead to strongly overestimated performances. In this setting, models learn solely from sequence similarities and node degrees. When data leakage is avoided by minimizing sequence similarities between training and test set, performances become random. Moreover, baseline models directly leveraging sequence similarity and network topology show good performances at a fraction of the computational cost. Thus, we advocate that any improvements should be reported relative to baseline methods in the future. Our findings suggest that predicting protein-protein interactions remains an unsolved task for proteins showing little sequence similarity to previously studied proteins, highlighting that further experimental research into the "dark" protein interactome and better computational methods are needed.

bioinformatics↗

bioRxiv2022

TF-Prioritizer: a java pipeline to prioritize condition-specific transcription factors

BackgroundEukaryotic gene expression is controlled by cis-regulatory elements (CREs), including promoters and enhancers, which are bound by transcription factors (TFs). Differential expression of TFs and their binding affinity at putative CREs determine tissue- and developmental-specific transcriptional activity. Consolidating genomic data sets can offer further insights into the accessibility of CREs, TF activity, and, thus, gene regulation. However, the integration and analysis of multi-modal data sets are hampered by considerable technical challenges. While methods for highlighting differential TF activity from combined chromatin state data (e.g., ChIP-seq, ATAC-seq, or DNase-seq) and RNA-seq data exist, they do not offer convenient usability, have limited support for large-scale data processing, and provide only minimal functionality for visually interpreting results. ResultsWe developed TF-Prioritizer, an automated pipeline that prioritizes condition-specific TFs from multi-modal data and generates an interactive web report. We demonstrated its potential by identifying known TFs along with their target genes, as well as previously unreported TFs active in lactating mouse mammary glands. Additionally, we studied a variety of ENCODE data sets for cell lines K562 and MCF-7, including twelve histone modification ChIP-seq as well as ATAC-seq and DNase-seq datasets, where we observe and discuss assay-specific differences. ConclusionTF-Prioritizer accepts ATAC-seq, DNase-seq, or ChIP-seq and RNA-seq data as input and identifies TFs with differential activity, thus offering an understanding of genome-wide gene regulation, potential pathogenesis, and therapeutic targets in biomedical research.

bioinformatics↗

bioRxiv2022

Inference of differential gene regulatory networks from gene expression data using boosted differential trees

Diseases can be caused by molecular perturbations that induce specific changes in regulatory interactions and their coordinated expression, also referred to as network rewiring. However, the detection of complex changes in regulatory connections remains a challenging task and would benefit from the development of novel non-parametric approaches. We developed a new ensemble method called BoostDiff (boosted differential regression trees) to infer a differential network discriminating between two conditions. BoostDiff builds an adaptively boosted (AdaBoost) ensemble of differential trees with respect to a target condition. To build the differential trees, we propose differential variance improvement as a novel splitting criterion. Variable importance measures derived from the resulting models are used to reflect changes in gene expression predictability and to build the output differential networks. BoostDiff outperforms existing differential network methods on simulated data evaluated in two different complexity settings. We then demonstrate the power of our approach when applied to real transcriptomics data in COVID-19 and Crohns disease. BoostDiff identifies context-specific networks that are enriched with genes of known disease-relevant pathways and complements standard differential expression analyses. BoostDiff is available at https://github.com/gihannagalindez/boostdiff_inference. Author SummaryGene regulatory networks, which comprise the collection of regulatory relationships between transcription factors and their target genes, are important for controlling various molecular processes. Diseases can induce perturbations in normal gene co-expression patterns in these networks. Detecting differentially co-expressed or rewired edges between disease and healthy biological states can be thus useful for investigating the link between specific disease-associated molecular alterations and phenotype. We developed BoostDiff (boosted differential trees), an ensemble method to derive differential networks between two biological contexts. Our approach applies a boosting scheme using differential trees as base learner. A differential tree is a new tree structure that is built from two expression datasets using a splitting criterion called the differential variance improvement. The resulting BoostDiff model learns the most differentially predictive features which are then used to build the directed differential networks. BoostDiff outperforms other differential network methods on simulated data and outputs more biologically meaningful results when evaluated on real transcriptomics datasets. BoostDiff can be applied to gene expression data to reveal new disease mechanisms or identify potential therapeutic targets.

bioinformatics↗