Archives

  • 2018-07
  • 2018-10
  • 2018-11
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • 2022-02
  • 2022-03
  • 2022-04
  • 2022-05
  • 2022-06
  • 2022-07
  • 2022-08
  • 2022-09
  • 2022-10
  • 2022-11
  • 2022-12
  • 2023-01
  • 2023-02
  • 2023-03
  • 2023-04
  • 2023-05
  • 2023-06
  • 2023-07
  • 2023-08
  • 2023-09
  • 2023-10
  • 2023-11
  • 2023-12
  • 2024-01
  • 2024-02
  • 2024-03
  • 2024-04

    • Here we present an algorithm KeyGenes

    2018-10-24

    Here we present an algorithm, KeyGenes, that we have used on NGS data extracted from tissues of 21 different human fetal organs, both embryonic and extraembryonic (plus the maternal endometrium), from the first and second trimesters of development to determine a panel of classifier genes that would be sufficient to confer identity to each fetal organ analyzed with high confidence. We showed that the developmental classifier genes selected were largely sufficient to predict the identity of their adult organ counterparts, even when using different types of platforms (NGS and microarray). Most importantly, as proof of concept, we challenged KeyGenes to identify a series of tissues using either recently published or our own NGS datasets. These included the following: (1) hPSCs differentiated to derivatives of the three germ lineages, namely, endoderm (pancreas), buy SR 3576 (brain), and mesoderm (heart); (2) tissue organoids (intestine); and (3) human fetal and adult organs/tissues. In all cases, KeyGenes accurately predicted tissue origin and, furthermore, we could use KeyGenes to assign a developmentally equivalent stage. KeyGenes is an easy-to-use, flexible, and expandable tool that can be applied to identify stem cell derivatives, when common marker profiles have been insufficiently informative, and provide benchmarking for protocols designed to promote maturation of stem cell derivatives in culture. KeyGenes is available at http://www.keygenes.nl.
    Results
    Discussion We have shown here through detailed genomic analysis that human organs and tissues retain a transcriptional signature from W9 until adulthood even though each organ is composed of multiple progenitor cell types that mature over time. It was remarkable that the transcriptional expression profile of a set of less than 100 genes was sufficient to identify 21 different human fetal organs/tissues (plus maternal endometrium) and 18 adult human organs. These classifier genes that we identified not only included genes involved in transcription regulation but also genes that define cellular shape and metabolism. Moreover, some of the classifier genes were lncRNAs and asRNAs, highlighting the regulatory importance of this class of genes (Washietl et al., 2014). This was notably illustrated by NPPA-AS1, which is thought to regulate NPPA expression, a gene that encodes atrial natriuretic factor and is involved in heart development and chamber specification (Annilo et al., 2009; Houweling et al., 2005). This identification also underscores one of the advantages of KeyGenes, developed to compare data to an NGS (fetal) training set, in contrast to existing algorithms (Cahan et al., 2014; Hwang et al., 2011; Morris et al., 2014) that compare data to networks deduced from microarray datasets, which contain a fixed set of probes with low representation of non-coding RNAs such as lncRNAs and asRNAs. In our case, three of the four lncRNAs identified as fetal classifier genes (RP13-49I15.5, NPPA-AS1, and LINC00514) were not present in the microarray adult dataset, but were present in the NGS adult dataset and identified there as well as classifier genes for predictions. The transcriptional human fetal atlas dataset presented here, even though limited in number of samples and organs analyzed, is an unique resource that will provide a deeper understanding of the signaling cascades and molecular dynamics during human development that lead into the maturation of progenitor cells within each human organ. The human fetal NGS dataset was paramount to the development and validation of our prediction algorithm KeyGenes, and has proven sufficient as training set to identify both human adult organs (using NGS and microarray data) as well as several differentiated derivatives of hPSCs. All hPSC derivatives expressed genes identified as fetal classifier genes (and many helper classifier genes) of the specific tissue to which they were claimed to have differentiated (Figure S5). Comparing differentiated derivatives of hPSCs to human adult gene expression data is important, but can be misleading, as it compares immature cells with far later stages of development when entirely different physiological parameters have affected cell behavior. Our fetal dataset, used alone or in combination with, for example, adult data as a training set, provided an efficient way to assess progression of hPSC differentiation.