Deep learning analysis of images and text unfolds new horizons in medicine. However, analysis of transcriptomic data, the cause of biological and pathological changes, is hampered by structural ...complexity distinctive from images and text. Here we conduct unsupervised training on more than 20,000 human normal and tumor transcriptomic data and show that the resulting Deep-Autoencoder, DeepT2Vec, has successfully extracted informative features and embedded transcriptomes into 30-dimensional Transcriptomic Feature Vectors (TFVs). We demonstrate that the TFVs could recapitulate expression patterns and be used to track tissue origins. Trained on these extracted features only, a supervised classifier, DeepC, can effectively distinguish tumors from normal samples with an accuracy of 90% for Pan-Cancer and reach an average 94% for specific cancers. Training on a connected network, the accuracy is further increased to 96% for Pan-Cancer. Together, our study shows that deep learning with autoencoder is suitable for transcriptomic analysis, and DeepT2Vec could be successfully applied to distinguish cancers, normal tissues, and other potential traits with limited samples.
We describe an mRNA profiling technique for determining differential gene expression that utilizes, but does not require, prior knowledge of gene sequences. This method permits high-throughput ...reproducible detection of most expressed sequences with a sensitivity of greater than 1 part in 100,000. Gene identification by database query of a restriction endonuclease fingerprint, confirmed by competitive PCR using gene-specific oligonucleotides, facilitates gene discovery by minimizing isolation procedures. This process, called GeneCalling, was validated by analysis of the gene expression profiles of normal and hypertrophic rat hearts following in vivo pressure overload.
Celotno besedilo
Dostopno za:
DOBA, IJS, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Assessment of tumor infiltrating lymphocytes (TILs) as a prognostic variable in melanoma has not seen broad adoption due to lack of standardization. Automation could represent a solution. Here, using ...open source software, we build an algorithm for image-based automated assessment of TILs on hematoxylin-eosin stained sections in melanoma. Using a retrospective collection of 641 melanoma patients comprising four independent cohorts; one training set (N = 227) and three validation cohorts (N = 137, N = 201, N = 76) from 2 institutions, we show that the automated TIL scoring algorithm separates patients into favorable and poor prognosis cohorts, where higher TILs scores were associated with favorable prognosis. In multivariable analyses, automated TIL scores show an independent association with disease-specific overall survival. Therefore, the open source, automated TIL scoring is an independent prognostic marker in melanoma. With further study, we believe that this algorithm could be useful to define a subset of patients that could potentially be spared immunotherapy.
•Both endogenous and exogenous mutational processes contribute to LUSC mutagenesis.•Variants contributing to the cancer phenotype are associated with aging, tobacco smoking, and APOBEC ...mutagenesis.•Variants of high effect in TP53 are attributable to signatures associated with tobacco smoking.•Variants of high effect in PIK3CA and NFE2L2 are attributable to signatures associated with APOBEC activity.•APOBEC3B effectively deaminates an NFE2L2 oligonucleotide sequence in vitro.
Lung squamous-cell carcinoma originates as a consequence of oncogenic molecular variants arising from diverse mutagenic processes such as tobacco, defective homologous recombination, aging, and cytidine deamination by APOBEC proteins. Only some of the many variants generated by these processes actually contribute to tumorigenesis. Therefore, molecular investigation of mutagenic processes such as cytidine deamination by APOBEC should also determine whether the mutations produced by these processes contribute substantially to the growth and survival of cancer. Here, we determine the processes that gave rise to mutations of 681 lung squamous-cell carcinomas, and quantify the probability that each mutation was the product of each process. We then calculate the contribution of each mutation to increases in cellular proliferation and survival. We performed in vitro experiments to determine cytidine deamination activity of APOBEC3B against oligonucleotides corresponding with genomic sequences that give rise to variants of high cancer effect size. The largest APOBEC-related cancer effects are attributable to mutations in PIK3CA and NFE2L2. We demonstrate that APOBEC effectively deaminates NFE2L2 at the locations that confer high cancer effect. Overall, we demonstrate that APOBEC activity can lead to mutations in NFE2L2 that have large contributions to cancer cell growth and survival, and that NFE2L2 is an attractive potential target for therapeutic intervention.
Short-chain fatty acids (SCFAs) butyrate, propionate, and acetate produced during fiber fermentation promote colonic differentiation and can reverse or suppress neoplastic progression. We sought to ...identify candidate genes responsible for SCFA activity on colonocytes and to compare the relative activities of independent SCFAs. cDNA was generated from polyA
+ mRNA isolated from control Caco-2 cells and cells treated with equimolar butyrate, propionate, and acetate. GeneCalling, a restriction-based differential mRNA expression platform linked to a DNA sequence database lookup, was applied. A total of 30,000 individual genetic sequences were analyzed for differential expression among the three SCFAs. Differentially expressed peaks corresponding to cancer-related genes were isolated, sequenced, and cross-referenced to the GenBank human database. Gene identities were independently confirmed by oligonucleotide poisoning. More than 1000 gene fragments were identified as being substantially modulated in expression by butyrate. Butyrate tended to have the most pronounced effects and acetate the least. Five fragments selected for further study were fully sequenced and proved 100% homologous with human sequences for clusterin, amyloid precursor-like protein 2, and caudal homeobox 2 protein, not previously known to be modulated by SCFAs. In each case, a similar order of potency for the three SCFAs studied was observed. The common SCFAs appear to exert different effects. This study suggests the diversity of the SCFA response at the molecular level and facilitates identifying genes important in the biologic activity of dietary fiber.
Expression pharmacogenomics includes differential gene expression (DGE) profiling of drug responses in model systems to generate a set of differentially modulated drug-responsive genes which can ...serve as a surrogate measure for drug action. In this manner, expression pharmacogenomics bridges the fields of genomics and medicinal chemistry. Additionally, modulated genes can be organized into metabolic and signaling pathways that highlight the mechanism of drug activity in a selected tissue. Here, we describe the application of expression pharmacogenomics to characterize a drug response in the clinically relevant in vivo model, the Sprague-Dawley rat. Following oral dosing of rats with GW9578, a novel synthetic peroxisome proliferator activated receptor alpha (PPAR alpha) ligand indicated for lipid disorders, we applied GeneCalling, a differential mRNA transcript profiling technique, to rat liver cDNA. Following GW9578 treatment, 2.4% of the rat liver genes were differentially expressed. We confirmed the sequence identity of 50 distinctly modulated genes. DGE was observed among genes representative of at least six discrete metabolic pathways. Furthermore, we observed up-regulation of 20 genes involved in mitochondrial, peroxisomal and microsomal fatty acid oxidation, consistent with molecular biological and clinical data indicating PPAR alpha ligand principal efficacy to be through increasing fatty acid metabolism. Those pathways regulated in our study that are potentially contributory to target effect, non-target adverse effects, or of unknown consequence include xenobiotic detoxification and steroid modification. Finally, comprehensive drug response profiling can lead to the serendipitous discovery of novel disease indications. In this case, these results suggest a potential novel indication for GW9578 in the treatment of X-linked adrenoleukodystrophy. We have shown, therefore, that the organization of DGE results into metabolic and signaling pathways can elucidate mechanisms of pharmacologically desired (i.e., efficacious) and, where appropriate, undesired (i.e., potentially deleterious) effects.
Expression pharmacogenomics includes differential gene expression (DGE) profiling of drug responses in model systems to generate a set of differentially modulated drug-responsive genes which can ...serve as a surrogate measure for drug action. In this manner, expression pharmacogenomics bridges the fields of genomics and medicinal chemistry. Additionally, modulated genes can be organized into metabolic and signaling pathways that highlight the mechanism of drug activity in a selected tissue. Here, we describe the application of expression pharmacogenomics to characterize a drug response in the clinically relevant in vivo model, the Sprague-Dawley rat. Following oral dosing of rats with GW9578, a novel synthetic peroxisome proliferator activated receptor alpha (PPAR Delta *a) ligand indicated for lipid disorders, we applied GeneCalling, a differential mRNA transcript profiling technique, to rat liver cDNA. Following GW9578 treatment, 2.4% of the rat liver genes were differentially expressed. We confirmed the sequence identity of 50 distinctly modulated genes. DGE was observed among genes representative of at least six discrete metabolic pathways. Furthermore, we observed up-regulation of 20 genes involved in mitochondrial, peroxisomal and microsomal fatty acid oxidation, consistent with molecular biological and clinical data indicating PPAR Delta *a ligand principal efficacy to be through increasing fatty acid metabolism. Those pathways regulated in our study that are potentially contributory to target effect, non-target adverse effects, or of unknown consequence include xenobiotic detoxification and steroid modification. Finally, comprehensive drug response profiling can lead to the serendipitous discovery of novel disease indications. In this case, these results suggest a potential novel indication for GW9578 in the treatment of X-linked adrenoleukodystrophy. We have shown, therefore, that the organization of DGE results into metabolic and signaling pathways can elucidate mechanisms of pharmacologically desired (i.e., efficacious) and, where appropriate, undesired (i.e., potentially deleterious) effects.
PDF
