OmicsFootPrint: a framework to integrate and interpret multi-omics data using circular images and deep neural networks.

The OmicsFootPrint framework addresses the need for advanced multi-omics data analysis methodologies by transforming data into intuitive two-dimensional circular images and facilitating the interpretation of complex diseases. Utilizing Deep Neural Networks and incorporating the SHapley Additive exPlanations (SHAP) algorithm, the framework enhances model interpretability. Tested with The Cancer Genome Atlas (TCGA) data, OmicsFootPrint effectively classified lung and breast cancer subtypes, achieving high Area Under Curve (AUC) scores - 0.98±0.02 for lung cancer subtype differentiation, 0.83±0.07 for breast cancer PAM50 subtypes, and successfully distinguishe between invasive lobular and ductal carcinomas in breast cancer, showcasing its robustness. It also demonstrated notable performance in predicting drug responses in cancer cell lines, with a median AUC of 0.74, surpassing existing algorithms. Furthermore, its effectiveness persists even with reduced training sample sizes. OmicsFootPrint marks an enhancement in multi-omics research, offering a novel, efficient, and interpretable approach that contributes to a deeper understanding of disease mechanisms.

Artificial intelligence of arterial Doppler waveforms to predict major adverse outcomes among patients with diabetes mellitus.

OBJECTIVE: Patients with diabetes mellitus (DM) are at increased risk for peripheral artery disease (PAD) and its complications. Arterial calcification and non-compressibility may limit test interpretation in this population. Developing tools capable of identifying PAD and predicting major adverse cardiac event (MACE) and limb event (MALE) outcomes among patients with DM would be clinically useful. Deep neural network analysis of resting Doppler arterial waveforms was used to detect PAD among patients with DM and to identify those at greatest risk for major adverse outcome events. METHODS: Consecutive patients with DM undergoing lower limb arterial testing (April 1, 2015-December 30, 2020) were randomly allocated to training, validation, and testing subsets (60%, 20%, and 20%). Deep neural networks were trained on resting posterior tibial arterial Doppler waveforms to predict all-cause mortality, MACE, and MALE at 5 years using quartiles based on the distribution of the prediction score. RESULTS: Among 11,384 total patients, 4211 patients with DM met study criteria (mean age, 68.6 ± 11.9 years; 32.0% female). After allocating the training and validation subsets, the final test subset included 856 patients. During follow-up, there were 262 deaths, 319 MACE, and 99 MALE. Patients in the upper quartile of prediction based on deep neural network analysis of the posterior tibial artery waveform provided independent prediction of death (hazard ratio [HR], 3.58; 95% confidence interval [CI], 2.31-5.56), MACE (HR, 2.06; 95% CI, 1.49-2.91), and MALE (HR, 13.50; 95% CI, 5.83-31.27). CONCLUSIONS: An artificial intelligence enabled analysis of a resting Doppler arterial waveform permits identification of major adverse outcomes including all-cause mortality, MACE, and MALE among patients with DM.

LPCAT1-TERT fusions are uniquely recurrent in epithelioid trophoblastic tumors and positively regulate cell growth.

Gestational trophoblastic disease (GTD) is a heterogeneous group of lesions arising from placental tissue. Epithelioid trophoblastic tumor (ETT), derived from chorionic-type trophoblast, is the rarest form of GTD with only approximately 130 cases described in the literature. Due to its morphologic mimicry of epithelioid smooth muscle tumors and carcinoma, ETT can be misdiagnosed. To date, molecular characterization of ETTs is lacking. Furthermore, ETT is difficult to treat when disease spreads beyond the uterus. Here using RNA-Seq analysis in a cohort of ETTs and other gestational trophoblastic lesions we describe the discovery of LPCAT1-TERT fusion transcripts that occur in ETTs and coincide with underlying genomic deletions. Through cell-growth assays we demonstrate that LPCAT1-TERT fusion proteins can positively modulate cell proliferation and therefore may represent future treatment targets. Furthermore, we demonstrate that TERT upregulation appears to be a characteristic of ETTs, even in the absence of LPCAT1-TERT fusions, and that it appears linked to copy number gains of chromosome 5. No evidence of TERT upregulation was identified in other trophoblastic lesions tested, including placental site trophoblastic tumors and placental site nodules, which are thought to be the benign chorionic-type trophoblast counterpart to ETT. These findings indicate that LPCAT1-TERT fusions and copy-number driven TERT activation may represent novel markers for ETT, with the potential to improve the diagnosis, treatment, and outcome for women with this rare form of GTD.

Indel sensitive and comprehensive variant/mutation detection from RNA sequencing data for precision medicine.

BACKGROUND: RNA-seq is the most commonly used sequencing application. Not only does it measure gene expression but it is also an excellent media to detect important structural variants such as single nucleotide variants (SNVs), insertion/deletion (Indels) or fusion transcripts. However, detection of these variants is challenging and complex from RNA-seq. Here we describe a sensitive and accurate analytical pipeline which detects various mutations at once for translational precision medicine. METHODS: The pipeline incorporates most sensitive aligners for Indels in RNA-Seq, the best practice for data preprocessing and variant calling, and STAR-fusion is for chimeric transcripts. Variants/mutations are annotated, and key genes can be extracted for further investigation and clinical actions. Three datasets were used to evaluate the performance of the pipeline for SNVs, indels and fusion transcripts. RESULTS: For the well-defined variants from NA12878 by GIAB project, about 95% and 80% of sensitivities were obtained for SNVs and indels, respectively, in matching RNA-seq. Comparison with other variant specific tools showed good performance of the pipeline. For the lung cancer dataset with 41 known and oncogenic mutations, 39 were detected by the pipeline with STAR aligner and all by the GSNAP aligner. An actionable EML4 and ALK fusion was also detected in one of the tumors, which also demonstrated outlier ALK expression. For 9 fusions spiked-into RNA-seq libraries with different concentrations, the pipeline was able to detect all in unfiltered results although some at very low concentrations may be missed when filtering was applied. CONCLUSIONS: The new RNA-seq workflow is an accurate and comprehensive mutation profiler from RNA-seq. Key or actionable mutations are reliably detected from RNA-seq, which makes it a practical alternative source for personalized medicine.

UClncR: Ultrafast and comprehensive long non-coding RNA detection from RNA-seq.

Long non-coding RNA (lncRNA) is a large class of gene transcripts with regulatory functions discovered in recent years. Many more are expected to be revealed with accumulation of RNA-seq data from diverse types of normal and diseased tissues. However, discovering novel lncRNAs and accurately quantifying known lncRNAs is not trivial from massive RNA-seq data. Herein we describe UClncR, an Ultrafast and Comprehensive lncRNA detection pipeline to tackle the challenge. UClncR takes standard RNA-seq alignment file, performs transcript assembly, predicts lncRNA candidates, quantifies and annotates both known and novel lncRNA candidates, and generates a convenient report for downstream analysis. The pipeline accommodates both un-stranded and stranded RNA-seq so that lncRNAs overlapping with other genes can be predicted and quantified. UClncR is fully parallelized in a cluster environment yet allows users to run samples sequentially without a cluster. The pipeline can process a typical RNA-seq sample in a matter of minutes and complete hundreds of samples in a matter of hours. Analysis of predicted lncRNAs from two test datasets demonstrated UClncR's accuracy and their relevance to sample clinical phenotypes. UClncR would facilitate researchers' novel lncRNA discovery significantly and is publically available at http://bioinformaticstools.mayo.edu/research/UClncR .

Indel detection from RNA-seq data: tool evaluation and strategies for accurate detection of actionable mutations.

Driver somatic mutations are a hallmark of a tumor that can be used for diagnosis and targeted therapy. Mutations are primarily detected from tumor DNA. As dynamic molecules of gene activities, transcriptome profiling by RNA sequence (RNA-seq) is becoming increasingly popular, which not only measures gene expression but also structural variations such as mutations and fusion transcripts. Although single-nucleotide variants (SNVs) can be easily identified from RNA-seq, intermediate long insertions/deletions (indels > 2 bases and less than sequence reads) cause significant challenges and are ignored by most RNA-seq analysis tools. This study evaluates commonly used RNA-seq analysis programs along with variant and somatic mutation callers in a series of data sets with simulated and known indels. The aim is to develop strategies for accurate indel detection. Our results show that the RNA-seq alignment is the most important step for indel identification and the evaluated programs have a wide range of sensitivity to map sequence reads with indels, from not at all to decently sensitive. The sensitivity is impacted by sequence read lengths. Most variant calling programs rely on hard evidence indels marked in the alignment and the programs with realignment may use soft-clipped reads for indel inferencing. Based on the observations, we have provided practical recommendations for indel detection when different RNA-seq aligners are used and demonstrated the best option with highly reliable results. With careful customization of bioinformatics algorithms, RNA-seq can be reliably used for both SNV and indel mutation detection that can be used for clinical decision-making.

Molecular Modeling and Functional Analysis of Exome Sequencing-Derived Variants of Unknown Significance Identify a Novel, Constitutively Active FGFR2 Mutant in Cholangiocarcinoma.

PURPOSE: Genomic testing has increased the quantity of information available to oncologists. Unfortunately, many identified sequence alterations are variants of unknown significance (VUSs), which thus limit the clinician's ability to use these findings to inform treatment. We applied a combination of in silico prediction and molecular modeling tools and laboratory techniques to rapidly define actionable VUSs. MATERIALS AND METHODS: Exome sequencing was conducted on 308 tumors from various origins. Most single nucleotide alterations within gene coding regions were VUSs. These VUSs were filtered to identify a subset of therapeutically targetable genes that were predicted with in silico tools to be altered in function by their variant sequence. A subset of receptor tyrosine kinase VUSs was characterized by laboratory comparison of each VUS versus its wild-type counterpart in terms of expression and signaling activity. RESULTS: The study identified 4,327 point mutations of which 3,833 were VUSs. Filtering for mutations in genes that were therapeutically targetable and predicted to affect protein function reduced these to 522VUSs of interest, including a large number of kinases. Ten receptortyrosine kinase VUSs were selected to explore in the laboratory. Of these, seven were found to be functionally altered. Three VUSs (FGFR2 F276C, FGFR4 R78H, and KDR G539R) showed increased basal or ligand-stimulated ERK phosphorylation compared with their wild-type counterparts, which suggests that they support transformation. Treatment of a patient who carried FGFR2 F276C with an FGFR inhibitor resulted in significant and sustained tumor response with clinical benefit. CONCLUSION: The findings demonstrate the feasibility of rapid identification of the biologic relevance of somatic mutations, which thus advances clinicians' ability to make informed treatment decisions.

Shiny FHIR: An Integrated Framework Leveraging Shiny R and HL7 FHIR to Empower Standards-Based Clinical Data Applications.

In this study, we describe our efforts in building a clinical statistics and analysis application platform using an emerging clinical data standard, HL7 FHIR, and an open source web application framework, Shiny. We designed two primary workflows that integrate a series of R packages to enable both patient-centered and cohort-based interactive analyses. We leveraged Shiny with R to develop interactive interfaces on FHIR-based data and used ovarian cancer study datasets as a use case to implement a prototype. Specifically, we implemented patient index, patient-centered data report and analysis, and cohort analysis. The evaluation of our study was performed by testing the adaptability of the framework on two public FHIR servers. We identify common research requirements and current outstanding issues, and discuss future enhancement work of the current studies. Overall, our study demonstrated that it is feasible to use Shiny for implementing interactive analysis on FHIR-based standardized clinical data.

Chromosome X genomic and epigenomic aberrations and clinical implications in breast cancer by base resolution profiling.

AIM: Abnormal inactivation or loss of inactivated X chromosome (Xi) is implicated in women's cancer. However, the underlying mechanisms and clinical relevance are little known. MATERIALS & METHODS: High-throughput sequencing was conducted on breast cancer cell lines for copy number, RNA expression and 5'-methylcytosine in ChrX. The results were examined in primary breast tumors. RESULTS & CONCLUSION: Breast cancer cells demonstrated reduced or total loss of hemimethylation. Most cell lines lost part or one of X chromosomes. Cell lines without ChrX loss were more active in gene expression. DNA methylation was corroborated with Xi control lincRNA XIST. Similar transcriptome and DNA methylation changes were observed in primary breast cancer datasets with clinical phenotype associations. Dramatic genomic and epigenomic changes in ChrX may be used for potential diagnostic or prognostic markers in breast cancer.

Nuclear factor of activated T-cell activity is associated with metastatic capacity in colon cancer.

Metastatic recurrence is the leading cause of cancer-related deaths in patients with colorectal carcinoma. To capture the molecular underpinnings for metastasis and tumor progression, we performed integrative network analysis on 11 independent human colorectal cancer gene expression datasets and applied expression data from an immunocompetent mouse model of metastasis as an additional filter for this biologic process. In silico analysis of one metastasis-related coexpression module predicted nuclear factor of activated T-cell (NFAT) transcription factors as potential regulators for the module. Cells selected for invasiveness and metastatic capability expressed higher levels of NFATc1 as compared with poorly metastatic and less invasive parental cells. We found that inhibition of NFATc1 in human and mouse colon cancer cells resulted in decreased invasiveness in culture and downregulation of metastasis-related network genes. Overexpression of NFATc1 significantly increased the metastatic potential of colon cancer cells, whereas inhibition of NFATc1 reduced metastasis growth in an immunocompetent mouse model. Finally, we found that an 8-gene signature comprising genes upregulated by NFATc1 significantly correlated with worse clinical outcomes in stage II and III colorectal cancer patients. Thus, NFATc1 regulates colon cancer cell behavior and its transcriptional targets constitute a novel, biologically anchored gene expression signature for the identification of colon cancers with high risk of metastatic recurrence.

PatternCNV: a versatile tool for detecting copy number changes from exome sequencing data.

MOTIVATION: Exome sequencing (exome-seq) data, which are typically used for calling exonic mutations, have also been utilized in detecting DNA copy number variations (CNVs). Despite the existence of several CNV detection tools, there is still a great need for a sensitive and an accurate CNV-calling algorithm with built-in QC steps, and does not require a paired reference for each sample. RESULTS: We developed a novel method named PatternCNV, which (i) accounts for the read coverage variations between exons while leveraging the consistencies of this variability across different samples; (ii) reduces alignment BAM files to WIG format and therefore greatly accelerates computation; (iii) incorporates multiple QC measures designed to identify outlier samples and batch effects; and (iv) provides a variety of visualization options including chromosome, gene and exon-level views of CNVs, along with a tabular summarization of the exon-level CNVs. Compared with other CNV-calling algorithms using data from a lymphoma exome-seq study, PatternCNV has higher sensitivity and specificity. AVAILABILITY AND IMPLEMENTATION: The software for PatternCNV is implemented using Perl and R, and can be used in Mac or Linux environments. Software and user manual are available at http://bioinformaticstools.mayo.edu/research/patterncnv/, and R package at https://github.com/topsoil/patternCNV/.

Adverse Drug Event-based Stratification of Tumor Mutations: A Case Study of Breast Cancer Patients Receiving Aromatase Inhibitors.

Adverse drug events (ADEs) are a critical factor for selecting cancer therapy options. The underlying molecular mechanisms of ADEs associated with cancer therapy drugs may overlap with their antineoplastic mechanisms; an aspect of toxicity. In the present study, we develop a novel knowledge-driven approach that provides an ADE-based stratification (ADEStrata) of tumor mutations. We demonstrate clinical utility of the ADEStrata approach through performing a case study of breast invasive carcinoma (BRCA) patients receiving aromatase inhibitors (AI) from The Cancer Genome Atlas (TCGA) (n=212), focusing on the musculoskeletal adverse events (MS-AEs). We prioritized somatic variants in a manner that is guided by MS-AEs codified as 6 Human Phenotype Ontology (HPO) terms. Pathway enrichment and hierarchical clustering of prioritized variants reveals clusters associated with overall survival. We demonstrated that the prediction of per-patient ADE propensity simultaneously identifies high-risk patients experiencing poor outcomes. In conclusion, the ADEStrata approach could produce clinically and biologically meaningful tumor subtypes that are potentially predictive of the drug response to the cancer therapy drugs.