TRAM (Transcriptome Mapper): database-driven creation and analysis of transcriptome maps from multiple sources

Zinc metabolism and its role in immunity status in subjects with trisomy 21: chromosomal dosage effect

Introduction

Trisomy 21 (T21), which causes Down syndrome (DS), is the most common chromosomal aneuploidy in humankind and includes different clinical comorbidities, among which the alteration of the immune system has a heavy impact on patient's lives. A molecule with an important role in immune response is zinc and it is known that its concentration is significantly lower in children with T21. Different hypotheses were made about this metabolic alteration and one of the reasons might be the overexpression of superoxide dismutase 1 (SOD1) gene, as zinc is part of the SOD1 active enzymatic center.

Methods

The aim of our work is to explore if there is a linear correlation between zinc level and immune cell levels measured in a total of 217 blood samples from subjects with T21. Furthermore, transcriptome map analyses were performed using Transcriptome Mapper (TRAM) software to investigate whether a difference in gene expression is detectable between subjects with T21 and euploid control group in tissues and cells involved in the immune response such as lymphoblastoid cells, thymus and white blood cells.

Results

Our results have confirmed the literature data stating that the blood zinc level in subjects with T21 is lower compared to the general population; in addition, we report that the T21/control zinc concentration ratio is 2:3, consistent with a chromosomal dosage effect due to the presence of three copies of chromosome 21. The transcriptome map analyses showed an alteration of some gene's expression which might explain low levels of zinc in the blood.

Discussion

Our data suggest that zinc level is not associated with the levels of immunity cells or proteins analyzed themselves and rather the main role of this ion might be played in altering immune cell function.

Gene Expression Studies in Down Syndrome: What Do They Tell Us about Disease Phenotypes?

Down syndrome is a well-studied aneuploidy condition in humans, which is associated with various disease phenotypes including cardiovascular, neurological, haematological and immunological disease processes. This review paper aims to discuss the research conducted on gene expression studies during fetal development. A descriptive review was conducted, encompassing all papers published on the PubMed database between September 1960 and September 2022. We found that in amniotic fluid, certain genes such as COL6A1 and DSCR1 were found to be affected, resulting in phenotypical craniofacial changes. Additionally, other genes such as GSTT1, CLIC6, ITGB2, C21orf67, C21orf86 and RUNX1 were also identified to be affected in the amniotic fluid. In the placenta, dysregulation of genes like MEST, SNF1LK and LOX was observed, which in turn affected nervous system development. In the brain, dysregulation of genes DYRK1A, DNMT3L, DNMT3B, TBX1, olig2 and AQP4 has been shown to contribute to intellectual disability. In the cardiac tissues, dysregulated expression of genes GART, ETS2 and ERG was found to cause abnormalities. Furthermore, dysregulation of XIST, RUNX1, SON, ERG and STAT1 was observed, contributing to myeloproliferative disorders. Understanding the differential expression of genes provides insights into the genetic consequences of DS. A better understanding of these processes could potentially pave the way for the development of genetic and pharmacological therapies.

The transcriptome profile of human trisomy 21 blood cells

BACKGROUND

Trisomy 21 (T21) is a genetic alteration characterised by the presence of an extra full or partial human chromosome 21 (Hsa21) leading to Down syndrome (DS), the most common form of intellectual disability (ID). It is broadly agreed that the presence of extra genetic material in T21 gives origin to an altered expression of genes located on Hsa21 leading to DS phenotype. The aim of this study was to analyse T21 and normal control blood cell gene expression profiles obtained by total RNA sequencing (RNA-Seq).

RESULTS

The results were elaborated by the TRAM (Transcriptome Mapper) software which generated a differential transcriptome map between human T21 and normal control blood cells providing the gene expression ratios for 17,867 loci. The obtained gene expression profiles were validated through real-time reverse transcription polymerase chain reaction (RT-PCR) assay and compared with previously published data. A post-analysis through transcriptome mapping allowed the identification of the segmental (regional) variation of the expression level across the whole genome (segment-based analysis of expression). Interestingly, the most over-expressed genes encode for interferon-induced proteins, two of them (MX1 and MX2 genes) mapping on Hsa21 (21q22.3). The altered expression of genes involved in mitochondrial translation and energy production also emerged, followed by the altered expression of genes encoding for the folate cycle enzyme, GART, and the folate transporter, SLC19A1.

CONCLUSIONS

The alteration of these pathways might be linked and involved in the manifestation of ID in DS.

The Role of Artificial Intelligence in the Diagnosis and Prognosis of Renal Cell Tumors

The increasing availability of molecular data provided by next-generation sequencing (NGS) techniques is allowing improvement in the possibilities of diagnosis and prognosis in renal cancer. Reliable and accurate predictors based on selected gene panels are urgently needed for better stratification of renal cell carcinoma (RCC) patients in order to define a personalized treatment plan. Artificial intelligence (AI) algorithms are currently in development for this purpose. Here, we reviewed studies that developed predictors based on AI algorithms for diagnosis and prognosis in renal cancer and we compared them with non-AI-based predictors. Comparing study results, it emerges that the AI prediction performance is good and slightly better than non-AI-based ones. However, there have been only minor improvements in AI predictors in terms of accuracy and the area under the receiver operating curve (AUC) over the last decade and the number of genes used had little influence on these indices. Furthermore, we highlight that different studies having the same goal obtain similar performance despite the fact they use different discriminating genes. This is surprising because genes related to the diagnosis or prognosis are expected to be tumor-specific and independent of selection methods and algorithms. The performance of these predictors will be better with the improvement in the learning methods, as the number of cases increases and by using different types of input data (e.g., non-coding RNAs, proteomic and metabolic). This will allow for more precise identification, classification and staging of cancerous lesions which will be less affected by interpathologist variability.

Sex-Specific Transcriptome Differences in Human Adipose Mesenchymal Stem Cells

In humans, sexual dimorphism can manifest in many ways and it is widely studied in several knowledge fields. It is increasing the evidence that also cells differ according to sex, a correlation still little studied and poorly considered when cells are used in scientific research. Specifically, our interest is on the sex-related dimorphism on the human mesenchymal stem cells (hMSCs) transcriptome. A systematic meta-analysis of hMSC microarrays was performed by using the Transcriptome Mapper (TRAM) software. This bioinformatic tool was used to integrate and normalize datasets from multiple sources and allowed us to highlight chromosomal segments and genes differently expressed in hMSCs derived from adipose tissue (hADSCs) of male and female donors. Chromosomal segments and differentially expressed genes in male and female hADSCs resulted to be related to several processes as inflammation, adipogenic and neurogenic differentiation and cell communication. Obtained results lead us to hypothesize that the donor sex of hADSCs is a variable influencing a wide range of stem cell biologic processes. We believe that it should be considered in biologic research and stem cell therapy.

Transportin 3 (TNPO3) and related proteins in limb girdle muscular dystrophy D2 muscle biopsies: A morphological study and pathogenetic hypothesis

LGMD D2 is a disease caused by TNPO3 mutation. We describe the expression of TNPO3 and selected proteins, likely modified by TNPO3 mutation, in muscle biopsies of affected patients. We also aim to find other genes involved in pathways correlated to TNPO3. Our morphological study on LGMD D2 muscle described the expression of TNPO3 and SRSF1, a splicing factor transported by TNPO3. Moreover, we investigated some sarcomeric and nuclear proteins, likely altered by TNPO3 mutation. Through an in silico approach we tried to identify genes involved in pathways that include, besides TNPO3 and SRSF1, p62 and Murf-1, altered in LGMD D2. In patients' muscles TNPO3 appeared weaker and randomly organized, with sporadic cytoplasmic aggregates positive for TNPO3; both SRSF1 and sarcomeric alpha actinin showed a different expression, while there were no alterations in the expression of the nuclear proteins. The in silico study lead to identify five genes, all coding for proteins responsible for muscle contraction. Our data suggest a possible interference in the morphology and function of myofibrillar network by mutated TNPO3; these findings are supported by the in silico identification of genes involved in muscle contraction that could help to explain the pathogenic mechanisms of LGMD D2.

Dysregulation of NIPBL leads to impaired RUNX1 expression and haematopoietic defects

The transcription factor RUNX1, a pivotal regulator of HSCs and haematopoiesis, is a frequent target of chromosomal translocations, point mutations or altered gene/protein dosage. These modifications lead or contribute to the development of myelodysplasia, leukaemia or platelet disorders. A better understanding of how regulatory elements contribute to fine‐tune the RUNX1 expression in haematopoietic tissues could improve our knowledge of the mechanisms responsible for normal haematopoiesis and malignancy insurgence. The cohesin RAD21 was reported to be a regulator of RUNX1 expression in the human myeloid HL60 cell line and during primitive haematopoiesis in zebrafish. In our study, we demonstrate that another cohesin, NIPBL, exerts positive regulation of RUNX1 in three different contexts in which RUNX1 displays important functions: in megakaryocytes derived from healthy donors, in bone marrow samples obtained from adult patients with acute myeloid leukaemia and during zebrafish haematopoiesis. In this model, we demonstrate that alterations in the zebrafish orthologue nipblb reduce runx1 expression with consequent defects in its erythroid and myeloid targets such as gata1a and spi1b in an opposite way to rad21. Thus, also in the absence of RUNX1 translocation or mutations, additional factors such as defects in the expression of NIPBL might induce haematological diseases.

Reference quantitative transcriptome dataset for adult Caenorhabditis elegans

Caenorhabditis elegans is a nematode widely used in biology and genomics as a model organism. We provide an integrated, quantitative reference map for the transcriptome of whole, wild type Bristol N2 strain C. elegans worms. The map has been obtained by meta-analysis of 110 gene expression profiles available in Gene Expression Omnibus (GEO) repository and integrated using the computational biology tool Transcriptome Mapper (TRAM). Following probe assignment to the relative locus and intra- and inter-sample normalization (in particular using the scaled quantile method), a mean, consensus reference value is provided for 45,932 transcripts, along with standard deviation. Expression values are all mapped in the context of genomic coordinates. The map provides easy access to relationships among expression values of different genes in this standard condition, highlights genomic segments with relatively high over-/under-expression and may serve as a reference to test for gene expression variation for both individual genes and the whole transcriptome in specific biological conditions (e.g. mutated strains or differently grown worms).

Dataset of differential gene expression between total normal human thyroid and histologically normal thyroid adjacent to papillary thyroid carcinoma

This article contains further data and information from our published manuscript [1]. We aim to identify significant transcriptome alterations of total normal human thyroid vs. histologically normal thyroid adjacent to papillary thyroid carcinoma. We performed a systematic meta-analysis of all the available gene expression profiles for the whole organ also collecting gene expression data for the normal thyroid adjacent to papillary thyroid carcinoma. A differential quantitative transcriptome reference map was generated by using TRAM (Transcriptome Mapper) software able to combine, normalize and integrate a total of 35 datasets from total normal thyroid and 40 datasets from histologically normal thyroid adjacent to papillary thyroid carcinoma from different sources. This analysis identified genes and genome segments that significantly discriminated the two groups of samples. Differentially expressed genes were grouped and enrichment function analyses were performed identifying the main features of the differentially expressed genes between total normal thyroid and histologically normal thyroid adjacent to papillary thyroid carcinoma. The search for housekeeping genes retrieved 414 loci.

On the length, weight and GC content of the human genome

Objective

Basic parameters commonly used to describe genomes including length, weight and relative guanine-cytosine (GC) content are widely cited in absence of a primary source. By using updated data and original software we determined these values to the best of our knowledge as standard reference for the whole human nuclear genome, for each chromosome and for mitochondrial DNA. We also devised a method to calculate the relative GC content in the whole messenger RNA sequence set and in transcriptomes by multiplying the GC content of each gene by its mean expression level.

Results

The male nuclear diploid genome extends for 6.27 Gigabase pairs (Gbp), is 205.00 cm (cm) long and weighs 6.41 picograms (pg). Female values are 6.37 Gbp, 208.23 cm, 6.51 pg. The individual variability and the implication for the DNA informational density in terms of bits/volume were discussed. The genomic GC content is 40.9%. Following analysis in different transcriptomes and species, we showed that the greatest deviation was observed in the pathological condition analysed (trisomy 21 leukaemic cells) and in Caenorhabditis elegans. Our results may represent a solid basis for further investigation on human structural and functional genomics while also providing a framework for other genome comparative analysis.

Electronic supplementary material

The online version of this article (10.1186/s13104-019-4137-z) contains supplementary material, which is available to authorized users.

Sex-Specific Transcriptome Differences in Substantia Nigra Tissue: A Meta-Analysis of Parkinson's Disease Data

Parkinson’s disease (PD) is one of the most common progressive neurodegenerative diseases. Clinical and epidemiological studies indicate that sex differences, as well as genetic components and ageing, can influence the prevalence, age at onset and symptomatology of PD. This study undertook a systematic meta-analysis of substantia nigra microarray data using the Transcriptome Mapper (TRAM) software to integrate and normalize a total of 10 suitable datasets from multiple sources. Four different analyses were performed according to default parameters, to better define the segments differentially expressed between PD patients and healthy controls, when comparing men and women data sets. The results suggest a possible regulation of specific sex-biased systems in PD susceptibility. TRAM software allowed us to highlight the different activation of some genomic regions and loci involved in molecular pathways related to neurodegeneration and neuroinflammatory mechanisms.

Integrated Quantitative Transcriptome Maps of Human Trisomy 21 Tissues and Cells

Down syndrome (DS) is due to the presence of an extra full or partial chromosome 21 (Hsa21). The identification of genes contributing to DS pathogenesis could be the key to any rational therapy of the associated intellectual disability. We aim at generating quantitative transcriptome maps in DS integrating all gene expression profile datasets available for any cell type or tissue, to obtain a complete model of the transcriptome in terms of both expression values for each gene and segmental trend of gene expression along each chromosome. We used the TRAM (Transcriptome Mapper) software for this meta-analysis, comparing transcript expression levels and profiles between DS and normal brain, lymphoblastoid cell lines, blood cells, fibroblasts, thymus and induced pluripotent stem cells, respectively. TRAM combined, normalized, and integrated datasets from different sources and across diverse experimental platforms. The main output was a linear expression value that may be used as a reference for each of up to 37,181 mapped transcripts analyzed, related to both known genes and expression sequence tag (EST) clusters. An independent example in vitro validation of fibroblast transcriptome map data was performed through “Real-Time” reverse transcription polymerase chain reaction showing an excellent correlation coefficient (r = 0.93, p < 0.0001) with data obtained in silico. The availability of linear expression values for each gene allowed the testing of the gene dosage hypothesis of the expected 3:2 DS/normal ratio for Hsa21 as well as other human genes in DS, in addition to listing genes differentially expressed with statistical significance. Although a fraction of Hsa21 genes escapes dosage effects, Hsa21 genes are selectively over-expressed in DS samples compared to genes from other chromosomes, reflecting a decisive role in the pathogenesis of the syndrome. Finally, the analysis of chromosomal segments reveals a high prevalence of Hsa21 over-expressed segments over the other genomic regions, suggesting, in particular, a specific region on Hsa21 that appears to be frequently over-expressed (21q22). Our complete datasets are released as a new framework to investigate transcription in DS for individual genes as well as chromosomal segments in different cell types and tissues.

Network-based co-expression analysis for exploring the potential diagnostic biomarkers of metastatic melanoma

Metastatic melanoma is an aggressive skin cancer and is one of the global malignancies with high mortality and morbidity. It is essential to identify and verify diagnostic biomarkers of early metastatic melanoma. Previous studies have systematically assessed protein biomarkers and mRNA-based expression characteristics. However, molecular markers for the early diagnosis of metastatic melanoma have not been identified. To explore potential regulatory targets, we have analyzed the gene microarray expression profiles of malignant melanoma samples by co-expression analysis based on the network approach. The differentially expressed genes (DEGs) were screened by the EdgeR package of R software. A weighted gene co-expression network analysis (WGCNA) was used for the identification of DEGs in the special gene modules and hub genes. Subsequently, a protein-protein interaction network was constructed to extract hub genes associated with gene modules. Finally, twenty-four important hub genes (RASGRP2, IKZF1, CXCR5, LTB, BLK, LINGO3, CCR6, P2RY10, RHOH, JUP, KRT14, PLA2G3, SPRR1A, KRT78, SFN, CLDN4, IL1RN, PKP3, CBLC, KRT16, TMEM79, KLK8, LYPD3 and LYPD5) were treated as valuable factors involved in the immune response and tumor cell development in tumorigenesis. In addition, a transcriptional regulatory network was constructed for these specific modules or hub genes, and a few core transcriptional regulators were found to be mostly associated with our hub genes, including GATA1, STAT1, SP1, and PSG1. In summary, our findings enhance our understanding of the biological process of malignant melanoma metastasis, enabling us to identify specific genes to use for diagnostic and prognostic markers and possibly for targeted therapy.

Systematic large-scale meta-analysis identifies a panel of two mRNAs as blood biomarkers for colorectal cancer detection

Colorectal cancer (CRC) is the third most common cancer in the world. A significant survival rate is achieved if it is detected at an early stage. A whole blood screening test, without any attempt to isolate blood fractions, could be an important tool to improve early detection of colorectal cancer. We searched for candidate markers with a novel approach based on the Transcriptome Mapper (TRAM), aimed at identifying specific RNAs with the highest differential expression ratio between colorectal cancer tissue and normal blood samples. This tool permits a large-scale systematic meta-analysis of all available data obtained by microarray experiments. The targeting of RNA took into consideration that tumour phenotypic variation is associated with changes in the mRNA levels of genes regulating or affecting this variation.

A real time quantitative reverse transcription polymerase chain reaction (qRT- PCR) was applied to the validation of candidate markers in the blood of 67 patients and 67 healthy controls. The expression of genes: TSPAN8, LGALS4, COL1A2 and CEACAM6 resulted as being statistically different.

In particular ROC curves attested for TSPAN8 an AUC of 0.751 with a sensitivity of 83.6% and a specificity of 58.2% at a cut off of 10.85, while the panel of the two best genes showed an AUC of 0.861 and a sensitivity of 92.5% with a specificity of 67.2%.

Our preliminary study on a total of 134 subjects showed promising results for a blood screening test to be validated in a larger cohort with the staging stratification and in patients with other gastrointestinal diseases.

LGALS4, CEACAM6, TSPAN8, and COL1A2: Blood Markers for Colorectal Cancer-Validation in a Cohort of Subjects With Positive Fecal Immunochemical Test Result

BACKGROUND

A noninvasive blood test for the early detection of colorectal cancer (CRC) is highly required. We evaluated a panel of 4 mRNAs as putative markers of CRC.

MATERIALS AND METHODS

We tested LGALS4, CEACAM6, TSPAN8, and COL1A2, referred to as the CELTiC panel, using quantitative reverse transcription polymerase chain reaction, on subjects with positive fecal immunochemical test (FIT) results and undergoing colonoscopy. Using a nonparametric test and multinomial logistic model, FIT-positive subjects were compared with CRC patients and healthy individuals.

RESULTS

All the genes of the CELTiC panel displayed statistically significant differences between the healthy subjects (n = 67), both low-risk (n = 36) and high-risk/CRC (n = 92) subjects, and those in the negative-colonoscopy, FIT-positive group (n = 36). The multinomial logistic model revealed LGALS4 was the most powerful marker discriminating the 4 groups. When assessing the diagnostic values by analysis of the areas under the receiver operating characteristic curves (AUCs), the CELTiC panel reached an AUC of 0.91 (sensitivity, 79%; specificity, 94%) comparing normal subjects to low-risk subjects, and 0.88 (sensitivity, 75%; specificity, 87%) comparing normal and high-risk/CRC subjects. The comparison between the normal subjects and the negative-colonoscopy, FIT-positive group revealed an AUC of 0.93 (sensitivity, 82%; specificity, 97%).

CONCLUSION

The CELTiC panel could represent a useful tool for discriminating subjects with positive FIT findings and for the early detection of precancerous adenomatous lesions and CRC.

A molecular view of the normal human thyroid structure and function reconstructed from its reference transcriptome map

BACKGROUND

The thyroid is the earliest endocrine structure to appear during human development, and thyroid hormones are necessary for proper organism development, in particular for the nervous system and heart, normal growth and skeletal maturation. To date a quantitative, validated transcriptional atlas of the whole normal human thyroid does not exist and the availability of a detailed expression map might be an excellent occasion to investigate the many features of the thyroid transcriptome.

RESULTS

We present a view at the molecular level of the normal human thyroid histology and physiology obtained by a systematic meta-analysis of all the available gene expression profiles for the whole organ. A quantitative transcriptome reference map was generated by using the TRAM (Transcriptome Mapper) software able to combine, normalize and integrate a total of 35 suitable datasets from different sources thus providing a typical reference expression value for each of the 27,275 known, mapped transcripts obtained. The experimental in vitro validation of data was performed by "Real-Time" reverse transcription polymerase chain reaction showing an excellent correlation coefficient (r = 0.93) with data obtained in silico.

CONCLUSIONS

Our study provides a quantitative global reference portrait of gene expression in the normal human thyroid and highlights differential expression between normal human thyroid and a pool of non-thyroid tissues useful for modeling correlations between thyroidal gene expression and specific thyroid functions and diseases. The experimental in vitro validation supports the possible usefulness of the human thyroid transcriptome map as a reference for molecular studies of the physiology and pathology of this organ.

Systematic identification of human housekeeping genes possibly useful as references in gene expression studies

The ideal reference, or control, gene for the study of gene expression in a given organism should be expressed at a medium-high level for easy detection, should be expressed at a constant/stable level throughout different cell types and within the same cell type undergoing different treatments, and should maintain these features through as many different tissues of the organism. From a biological point of view, these theoretical requirements of an ideal reference gene appear to be best suited to housekeeping (HK) genes. Recent advancements in the quality and completeness of human expression microarray data and in their statistical analysis may provide new clues toward the quantitative standardization of human gene expression studies in biology and medicine, both cross- and within-tissue. The systematic approach used by the present study is based on the Transcriptome Mapper tool and exploits the automated reassignment of probes to corresponding genes, intra- and inter-sample normalization, elaboration and representation of gene expression values in linear form within an indexed and searchable database with a graphical interface recording quantitative levels of expression, expression variability and cross-tissue width of expression for more than 31,000 transcripts. The present study conducted a meta-analysis of a pool of 646 expression profile data sets from 54 different human tissues and identified actin γ 1 as the HK gene that best fits the combination of all the traditional criteria to be used as a reference gene for general use; two ribosomal protein genes, RPS18 and RPS27, and one aquaporin gene, POM121 transmembrane nucleporin C, were also identified. The present study provided a list of tissue- and organ-specific genes that may be most suited for the following individual tissues/organs: Adipose tissue, bone marrow, brain, heart, kidney, liver, lung, ovary, skeletal muscle and testis; and also provides in these cases a representative, quantitative portrait of the relative, typical gene-expression profile in the form of searchable database tables.

Meta-Analysis of Parkinson's Disease Transcriptome Data Using TRAM Software: Whole Substantia Nigra Tissue and Single Dopamine Neuron Differential Gene Expression

The understanding of the genetic basis of the Parkinson's disease (PD) and the correlation between genotype and phenotype has revolutionized our knowledge about the pathogenetic mechanisms of neurodegeneration, opening up exciting new therapeutic and neuroprotective perspectives. Genomic knowledge of PD is still in its early stages and can provide a good start for studies of the molecular mechanisms that underlie the gene expression variations and the epigenetic mechanisms that may contribute to the complex and characteristic phenotype of PD. In this study we used the software TRAM (Transcriptome Mapper) to analyse publicly available microarray data of a total of 151 PD patients and 130 healthy controls substantia nigra (SN) samples, to identify chromosomal segments and gene loci differential expression. In particular, we separately analyzed PD patients and controls data from post-mortem snap-frozen SN whole tissue and from laser microdissected midbrain dopamine (DA) neurons, to better characterize the specific DA neuronal expression profile associated with the late-stage Parkinson's condition. The default "Map" mode analysis resulted in 10 significantly over/under-expressed segments, mapping on 8 different chromosomes for SN whole tissue and in 4 segments mapping on 4 different chromosomes for DA neurons. In conclusion, TRAM software allowed us to confirm the deregulation of some genomic regions and loci involved in key molecular pathways related to neurodegeneration, as well as to provide new insights about genes and non-coding RNA transcripts not yet associated with the disease.

Integrated Transcriptome Map Highlights Structural and Functional Aspects of the Normal Human Heart

A systematic meta-analysis of the available gene expression profiling datasets for the whole normal human heart generated a quantitative transcriptome reference map of this organ. Transcriptome Mapper (TRAM) software integrated 32 gene expression profile datasets from different sources returning a reference value of expression for each of the 43,360 known, mapped transcripts assayed by any of the experimental platforms used in this regard. Main findings include the visualization at the gene and chromosomal levels of the classical description of the basic histology and physiology of the heart, the identification of suitable housekeeping reference genes, the analysis of stoichiometry of gene products, and the focusing on chromosome 21 genes, which are present in one excess copy in Down syndrome subjects, presenting cardiovascular defects in 30-40% of cases. Independent in vitro validation showed an excellent correlation coefficient (r = 0.98) with the in silico data. Remarkably, heart/non-cardiac tissue expression ratio may also be used to anticipate that effects of mutations will most probably affect or not the heart. The quantitative reference global portrait of gene expression in the whole normal human heart illustrates the structural and functional aspects of the whole organ and is a general model to understand the mechanisms underlying heart pathophysiology. J. Cell. Physiol. 232: 759-770, 2017. © 2016 Wiley Periodicals, Inc.

Systematic reanalysis of partial trisomy 21 cases with or without Down syndrome suggests a small region on 21q22.13 as critical to the phenotype

A 'Down Syndrome critical region' (DSCR) sufficient to induce the most constant phenotypes of Down syndrome (DS) had been identified by studying partial (segmental) trisomy 21 (PT21) as an interval of 0.6-8.3 Mb within human chromosome 21 (Hsa21), although its existence was later questioned. We propose an innovative, systematic reanalysis of all described PT21 cases (from 1973 to 2015). In particular, we built an integrated, comparative map from 125 cases with or without DS fulfilling stringent cytogenetic and clinical criteria. The map allowed to define or exclude as candidates for DS fine Hsa21 sequence intervals, also integrating duplication copy number variants (CNVs) data. A highly restricted DSCR (HR-DSCR) of only 34 kb on distal 21q22.13 has been identified as the minimal region whose duplication is shared by all DS subjects and is absent in all non-DS subjects. Also being spared by any duplication CNV in healthy subjects, HR-DSCR is proposed as a candidate for the typical DS features, the intellectual disability and some facial phenotypes. HR-DSCR contains no known gene and has relevant homology only to the chimpanzee genome. Searching for HR-DSCR functional loci might become a priority for understanding the fundamental genotype-phenotype relationships in DS.

Weighted gene co-expression network analysis reveals key genes involved in pancreatic ductal adenocarcinoma development

PURPOSE

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive malignancy. Up till now, the patient's prognosis remains poor which, among others, is due to the paucity of reliable early diagnostic biomarkers. In the past, candidate diagnostic biomarkers and therapeutic targets have been delineated from genes that were found to be differentially expressed in normal versus tumour samples. Recently, new systems biology approaches have been developed to analyse gene expression data, which may yield new biomarkers. As of yet, the weighted gene co-expression network analysis (WGCNA) tool has not been applied to PDAC microarray-based gene expression data.

METHODS

PDAC microarray-based gene expression datasets, listed in the Gene Expression Omnibus (GEO) database, were analysed. After pre-processing of the data, we built two final datasets, Normal and PDAC, encompassing 104 and 129 patient samples, respectively. Next, we constructed a weighted gene co-expression network and identified modules of co-expressed genes distinguishing normal from disease conditions. Functional annotations of the genes in these modules were carried out to highlight PDAC-associated molecular pathways and common regulatory mechanisms. Finally, overall survival analyses were carried out to assess the suitability of the genes identified as prognostic biomarkers.

RESULTS

Using WGCNA, we identified several key genes that may play important roles in PDAC. These genes are mainly related to either endoplasmic reticulum, mitochondrion or membrane functions, exhibit transferase or hydrolase activities and are involved in biological processes such as lipid metabolism or transmembrane transport. As a validation of the applied method, we found that some of the identified key genes (CEACAM1, MCU, VDAC1, CYCS, C15ORF52, TMEM51, LARP1 and ERLIN2) have previously been reported by others as potential PDAC biomarkers. Using overall survival analyses, we found that several of the newly identified genes may serve as biomarkers to stratify PDAC patients into low- and high-risk groups.

CONCLUSIONS

Using this new systems biology approach, we identified several genes that appear to be critical to PDAC development. As such, they may represent potential diagnostic biomarkers as well as therapeutic targets with clinical utility.

Identification of minimal eukaryotic introns through GeneBase, a user-friendly tool for parsing the NCBI Gene databank

We have developed GeneBase, a full parser of the National Center for Biotechnology Information (NCBI) Gene database, which generates a fully structured local database with an intuitive user-friendly graphic interface for personal computers. Features of all the annotated eukaryotic genes are accessible through three main software tables, including for each entry details such as the gene summary, the gene exon/intron structure and the specific Gene Ontology attributions. The structuring of the data, the creation of additional calculation fields and the integration with nucleotide sequences allow users to make many types of comparisons and calculations that are useful for data retrieval and analysis. We provide an original example analysis of the existing introns across all the available species, through which the classic biological problem of the ‘minimal intron’ may find a solution using available data. Based on all currently available data, we can define the shortest known eukaryotic GT-AG intron length, setting the physical limit at the 30 base pair intron belonging to the human MST1L gene. This ‘model intron’ will shed light on the minimal requirement elements of recognition used for conventional splicing functioning. Remarkably, this size is indeed consistent with the sum of the splicing consensus sequence lengths.

A quantitative transcriptome reference map of the normal human hippocampus

We performed an innovative systematic meta-analysis of 41 gene expression profiles of normal human hippocampus to provide a quantitative transcriptome reference map of it, i.e. a reference typical value of expression for each of the 30,739 known mapped and the 16,258 uncharacterized (unmapped) transcripts. For this aim, we used the software called TRAM (Transcriptome Mapper), which is able to generate transcriptome maps based on gene expression data from multiple sources. We also analyzed differential expression by comparing the hippocampus with the whole brain transcriptome map to identify a typical expression pattern of this subregion compared with the whole organ. Finally, due to the fact that the hippocampus is one of the main brain region to be severely affected in trisomy 21 (the best known genetic cause of intellectual disability), a particular attention was paid to the expression of chromosome 21 (chr21) genes. Data were downloaded from microarray databases, processed, and analyzed using TRAM software. Among the main findings, the most over-expressed loci in the hippocampus are the expressed sequence tag cluster Hs.732685 and the member of the calmodulin gene family CALM2. The tubulin folding cofactor B (TBCB) gene is the best gene at behaving like a housekeeping gene. The hippocampus vs. the whole brain differential transcriptome map shows the over-expression of LINC00114, a long non-coding RNA mapped on chr21. The hippocampus transcriptome map was validated in vitro by assaying gene expression through several magnitude orders by "Real-Time" reverse transcription polymerase chain reaction (RT-PCR). The highly significant agreement between in silico and experimental data suggested that our transcriptome map may be a useful quantitative reference benchmark for gene expression studies related to human hippocampus. Furthermore, our analysis yielded biological insights about those genes that have an intrinsic over-/under-expression in the hippocampus.

Integrated differential transcriptome maps of Acute Megakaryoblastic Leukemia (AMKL) in children with or without Down Syndrome (DS)

BACKGROUND

The incidence of Acute Megakaryoblastic Leukemia (AMKL) is 500-fold higher in children with Down Syndrome (DS) compared with non-DS children, but the relevance of trisomy 21 as a specific background of AMKL in DS is still an open issue. Several Authors have determined gene expression profiles by microarray analysis in DS and/or non-DS AMKL. Due to the rarity of AMKL, these studies were typically limited to a small group of samples.

METHODS

We generated integrated quantitative transcriptome maps by systematic meta-analysis from any available gene expression profile dataset related to AMKL in pediatric age. This task has been accomplished using a tool recently described by us for the generation and the analysis of quantitative transcriptome maps, TRAM (Transcriptome Mapper), which allows effective integration of data obtained from different experimenters, experimental platforms and data sources. This allowed us to explore gene expression changes involved in transition from normal megakaryocytes (MK, n=19) to DS (n=43) or non-DS (n=45) AMKL blasts, including the analysis of Transient Myeloproliferative Disorder (TMD, n=20), a pre-leukemia condition.

RESULTS

We propose a biological model of the transcriptome depicting progressive changes from MK to TMD and then to DS AMKL. The data indicate the repression of genes involved in MK differentiation, in particular the cluster on chromosome 4 including PF4 (platelet factor 4) and PPBP (pro-platelet basic protein); the gene for the mitogen-activated protein kinase MAP3K10 and the thrombopoietin receptor gene MPL. Moreover, comparing both DS and non-DS AMKL with MK, we identified three potential clinical markers of progression to AMKL: TMEM241 (transmembrane protein 241) was the most over-expressed single gene, while APOC2 (apolipoprotein C-II) and ZNF587B (zinc finger protein 587B) appear to be the most discriminant markers of progression, specifically to DS AMKL. Finally, the chromosome 21 (chr21) genes resulted to be the most over-expressed in DS and non-DS AMKL, as well as in TMD, pointing out a key role of chr21 genes in differentiating AMKL from MK.

CONCLUSIONS

Our study presents an integrated original model of the DS AMLK transcriptome, providing the identification of genes relevant for its pathophysiology which can potentially be new clinical markers.

A quantitative transcriptome reference map of the normal human brain

We performed an innovative systematic meta-analysis of 60 gene expression profiles of whole normal human brain, to provide a quantitative transcriptome reference map of it, i.e. a reference typical value of expression for each of the 39,250 known, mapped and 26,026 uncharacterized (unmapped) transcripts. To this aim, we used the software named Transcriptome Mapper (TRAM), which is able to generate transcriptome maps based on gene expression data from multiple sources. We also analyzed differential expression by comparing the brain transcriptome with those derived from human foetal brain gene expression, from a pool of human tissues (except the brain) and from the two normal human brain regions cerebellum and cerebral cortex, which are two of the main regions severely affected when cognitive impairment occurs, as happens in the case of trisomy 21. Data were downloaded from microarray databases, processed and analyzed using TRAM software and validated in vitro by assaying gene expression through several magnitude orders by 'real-time' reverse transcription polymerase chain reaction (RT-PCR). The excellent agreement between in silico and experimental data suggested that our transcriptome maps may be a useful quantitative reference benchmark for gene expression studies related to the human brain. Furthermore, our analysis yielded biological insights about those genes which have an intrinsic over-/under-expression in the brain, in addition offering a basis for the regional analysis of gene expression. This could be useful for the study of chromosomal alterations associated to cognitive impairment, such as trisomy 21, the most common genetic cause of intellectual disability.

Improving mRNA 5' coding sequence determination in the mouse genome

The incomplete determination of the mRNA 5' end sequence may lead to the incorrect assignment of the first AUG codon and to errors in the prediction of the encoded protein product. Due to the significance of the mouse as a model organism in biomedical research, we performed a systematic identification of coding regions at the 5' end of all known mouse mRNAs, using an automated expressed sequence tag (EST)-based approach which we have previously described. By parsing almost 4 million BLAT alignments we found 351 mouse loci, out of 20,221 analyzed, in which an extension of the mRNA 5' coding region was identified. Proof-of-concept confirmation was obtained by in vitro cloning and sequencing for Apc2 and Mknk2 cDNAs. We also generated a list of 16,330 mouse mRNAs where the presence of an in-frame stop codon upstream of the known start codon indicates completeness of the coding sequence at 5' end in the current form. Systematic searches in the main mouse genome databases and genome browsers showed that 82% of our results are original and have not been identified by their annotation pipelines. Moreover, the same information is not easily derivable from RNA-Seq data, due to short sequence length and laboriousness in building full-length transcript structures. In conclusion, our results improve the determination of full-length 5' coding sequences and might be useful in order to reduce errors when studying mouse gene structure and function in biomedical research.

Universal tight correlation of codon bias and pool of RNA codons (codonome): The genome is optimized to allow any distribution of gene expression values in the transcriptome from bacteria to humans

Codon bias is the phenomenon in which distinct synonymous codons are used with different frequencies. We define here the "codonome value" as the total number of codons present across all the expressed mRNAs in a given biological condition. We have developed the "CODONOME" software, which calculates the codon bias and, following integration with a gene expression profile, estimates the actual frequency of each codon at the transcriptome level (codonome bias) of a given tissue. Systematic analysis across different human tissues and multiple species shows a surprisingly tight correlation between the codon bias and the codonome bias. An aneuploidy and cancer condition such as that of Down Syndrome-related acute megakaryoblastic leukemia (DS-AMKL), does not appear to alter this relationship. The law of correlation between codon bias and codonome emerges as a property of the distribution and range of the number, sequence and expression level of the genes in a genome.

Genome-scale analysis of human mRNA 5' coding sequences based on expressed sequence tag (EST) database

The "5' end mRNA artifact" issue refers to the incorrect assignment of the first AUG codon in an mRNA, due to the incomplete determination of its 5' end sequence. We performed a systematic identification of coding regions at the 5' end of all human known mRNAs, using an automated expressed sequence tag (EST)-based approach. Following parsing of more than 7 million BLAT alignments, we found 477 human loci, out of 18,665 analyzed, in which an extension of the mRNA 5' coding region was identified. Proof-of-concept confirmation was obtained by in vitro cloning and sequencing for GNB2L1, QARS and TDP2 cDNAs, and the consequences for the functional studies of these loci are discussed. We also generated a list of 20,775 human mRNAs where the presence of an in-frame stop codon upstream of the known start codon indicates completeness of the coding sequence at 5' in the current form.

MAPT and PAICE: Tools for time series and single time point transcriptionist visualization and knowledge discovery

With the advent of next-generation sequencing, -omics fields such as transcriptomics have experienced increases in data throughput on the order of magnitudes. In terms of analyzing and visually representing these huge datasets, an intuitive and computationally tractable approach is to map quantified transcript expression onto biochemical pathways while employing datamining and visualization principles to accelerate knowledge discovery. We present two cross-platform tools: MAPT (Mapping and Analysis of Pathways through Time) and PAICE (Pathway Analysis and Integrated Coloring of Experiments), an easy to use analysis suite to facilitate time series and single time point transcriptomics analysis. In unison, MAPT and PAICE serve as a visual workbench for transcriptomics knowledge discovery, data-mining and functional annotation. Both PAICE and MAPT are two distinct but yet inextricably linked tools. The former is specifically designed to map EC accessions onto KEGG pathways while handling multiple gene copies, detection-call analysis, as well as UN/annotated EC accessions lacking quantifiable expression. The latter tool integrates PAICE datasets to drive visualization, annotation, and data-mining.

AVAILABILITY

The database is available for free at http://sourceforge.net/projects/paice/http://sourceforge.net/projects/mapt/