BioVDB: biological vector database for high-throughput gene expression meta-analysis.

High-throughput sequencing has created an exponential increase in the amount of gene expression data, much of which is freely, publicly available in repositories such as NCBI's Gene Expression Omnibus (GEO). Querying this data for patterns such as similarity and distance, however, becomes increasingly challenging as the total amount of data increases. Furthermore, vectorization of the data is commonly required in Artificial Intelligence and Machine Learning (AI/ML) approaches. We present BioVDB, a vector database for storage and analysis of gene expression data, which enhances the potential for integrating biological studies with AI/ML tools. We used a previously developed approach called Automatic Label Extraction (ALE) to extract sample labels from metadata, including age, sex, and tissue/cell-line. BioVDB stores 438,562 samples from eight microarray GEO platforms. We show that it allows for efficient querying of data using similarity search, which can also be useful for identifying and inferring missing labels of samples, and for rapid similarity analysis.

Methylation Array Signals are Predictive of Chronological Age Without Bisulfite Conversion.

DNA methylation data has been used to make "epigenetic clocks" which attempt to measure chronological and biological aging. These models rely on data derived from bisulfite-based measurements, which exploit a semi-selective deamination and a genomic reference to determine methylation states. Here, we demonstrate how another hallmark of aging, genomic instability, influences methylation measurements in both bisulfite sequencing and methylation arrays. We found that non-methylation factors lead to "pseudomethylation" signals that are both confounding of epigenetic clocks and uniquely age predictive. Quantifying these covariates in aging studies will be critical to building better clocks and designing appropriate studies of epigenetic aging.

Many chronological aging clocks can be found throughout the epigenome: Implications for quantifying biological aging.

Epigenetic alterations are a hallmark of aging and age-related diseases. Computational models using DNA methylation data can create "epigenetic clocks" which are proposed to reflect "biological" aging. Thus, it is important to understand the relationship between predictive clock sites and aging biology. To do this, we examined over 450,000 methylation sites from 9,699 samples. We found ~20% of the measured genomic cytosines can be used to make many different epigenetic clocks whose age prediction performance surpasses that of telomere length. Of these predictive sites, the average methylation change over a lifetime was small (~1.5%) and these sites were under-represented in canonical regions of epigenetic regulation. There was only a weak association between "accelerated" epigenetic aging and disease. We also compare tissue-specific and pan-tissue clock performance. This is critical to applying clocks both to new sample sets in basic research, as well as understanding if clinically available tissues will be feasible samples to evaluate "epigenetic aging" in unavailable tissues (e.g., brain). Despite the reproducible and accurate age predictions from DNA methylation data, these findings suggest they may have limited utility as currently designed in understanding the molecular biology of aging and may not be suitable as surrogate endpoints in studies of anti-aging interventions. Purpose-built clocks for specific tissues age ranges or phenotypes may perform better for their specific purpose. However, if purpose-built clocks are necessary for meaningful predictions, then the utility of clocks and their application in the field needs to be considered in that context.

MNEMONIC: MetageNomic Experiment Mining to create an OTU Network of Inhabitant Correlations.

BACKGROUND: The number of publicly available metagenomic experiments in various environments has been rapidly growing, empowering the potential to identify similar shifts in species abundance between different experiments. This could be a potentially powerful way to interpret new experiments, by identifying common themes and causes behind changes in species abundance. RESULTS: We propose a novel framework for comparing microbial shifts between conditions. Using data from one of the largest human metagenome projects to date, the American Gut Project (AGP), we obtain differential abundance vectors for microbes using experimental condition information provided with the AGP metadata, such as patient age, dietary habits, or health status. We show it can be used to identify similar and opposing shifts in microbial species, and infer putative interactions between microbes. Our results show that groups of shifts with similar effects on microbiome can be identified and that similar dietary interventions display similar microbial abundance shifts. CONCLUSIONS: Without comparison to prior data, it is difficult for experimentalists to know if their observed changes in species abundance have been observed by others, both in their conditions and in others they would never consider comparable. Yet, this can be a very important contextual factor in interpreting the significance of a shift. We've proposed and tested an algorithmic solution to this problem, which also allows for comparing the metagenomic signature shifts between conditions in the existing body of data.

OKN-007 Increases temozolomide (TMZ) Sensitivity and Suppresses TMZ-Resistant Glioblastoma (GBM) Tumor Growth.

Treatment of glioblastoma (GBM) remains a challenge using conventional chemotherapy, such as temozolomide (TMZ), and is often ineffective as a result of drug resistance. We have assessed a novel nitrone-based agent, OKN-007, and found it to be effective in decreasing tumor volumes and increasing survival in orthotopic GBM xenografts by decreasing cell proliferation and angiogenesis and increasing apoptosis. In this study, we assessed combining OKN-007 with TMZ in vivo in a human G55 GBM orthotopic xenograft model and in vitro in TMZ-resistant and TMZ-sensitive human GBM cell lines. For the in vivo studies, magnetic resonance imaging was used to assess tumor growth and vascular alterations. Percent animal survival was also determined. For the in vitro studies, cell growth, IC50 values, RNA-seq, RT-PCR, and ELISA were used to assess growth inhibition, possible mechanism-of actions (MOAs) associated with combined OKN-007 + TMZ versus TMZ alone, and gene and protein expression levels, respectively. Microarray analysis of OKN-007-treated rat F98 glioma tumors was also carried out to determine possible MOAs of OKN-007 in glioma-bearing animals either treated or not treated with OKN-007. OKN-007 seems to elicit its effect on GBM tumors via inhibition of tumorigenic TGF-ß1, which affects the extracellular matrix. When combined with TMZ, OKN-007 significantly increases percent survival, decreases tumor volumes, and normalizes tumor blood vasculature in vivo compared to untreated tumors and seems to affect TMZ-resistant GBM cells possibly via IDO-1, SUMO2, and PFN1 in vitro. Combined OKN-007 + TMZ may be a potentially potent treatment strategy for GBM patients.

ALE: automated label extraction from GEO metadata.

BACKGROUND: NCBI's Gene Expression Omnibus (GEO) is a rich community resource containing millions of gene expression experiments from human, mouse, rat, and other model organisms. However, information about each experiment (metadata) is in the format of an open-ended, non-standardized textual description provided by the depositor. Thus, classification of experiments for meta-analysis by factors such as gender, age of the sample donor, and tissue of origin is not feasible without assigning labels to the experiments. Automated approaches are preferable for this, primarily because of the size and volume of the data to be processed, but also because it ensures standardization and consistency. While some of these labels can be extracted directly from the textual metadata, many of the data available do not contain explicit text informing the researcher about the age and gender of the subjects with the study. To bridge this gap, machine-learning methods can be trained to use the gene expression patterns associated with the text-derived labels to refine label-prediction confidence. RESULTS: Our analysis shows only 26% of metadata text contains information about gender and 21% about age. In order to ameliorate the lack of available labels for these data sets, we first extract labels from the textual metadata for each GEO RNA dataset and evaluate the performance against a gold standard of manually curated labels. We then use machine-learning methods to predict labels, based upon gene expression of the samples and compare this to the text-based method. CONCLUSION: Here we present an automated method to extract labels for age, gender, and tissue from textual metadata and GEO data using both a heuristic approach as well as machine learning. We show the two methods together improve accuracy of label assignment to GEO samples.