Archives

  • 2018-07
  • 2020-03
  • 2020-07
  • 2020-08
  • 477-30-5 br Table br Healthy Homo sapiens genes

    2020-08-12


    Table 3
    Healthy Homo sapiens genes.
    Gene nature Gene length range Gene ID Gene name Block length
    Hydrophobic > 450 ATBFRUCT1 Glycosyl hydrolases family 32 protein 541
    ATCWINV1 Beta-fructofuranosidase 537
    GLB1 Galactosidase beta 1 678
    KDM1A Lysine demethylase 1A 686
    MYO1C Myosin IC 725
    cmeC Multidrug efflux pump protein CmeC 479
    TTHA1135 ba3-type cytochrome C oxidase polypeptide I 568
    acrB Multidrug efflux system protein 1057
    ECK0456 Multidrug efflux pump subunit AcrB 1057
    spr1652 Cell wall surface anchor family protein 648
    FSHMD1A Facioscapulohumeral muscular dystrophy 1A 802
    bamA Outer membrane protein assembly factor BamA 532
    Table 4
    Measured phase values (deg) for Homo sapiens's genes.
    Gene type Gene ID Frequency in Hz
    amino 477-30-5 in the pure hydrophilic gene chain. Resistance Rn is same for all hydrophobic and hydrophilic amino acids.
    Now the recurrence relations for gene chain consists of hydrophobic and hydrophilic residues both, are as follows:
    where Gn = Nn/Dn, Nn and Dn are the polynomials of degree n for both hydrophilic and hydrophobic genes. Therefore using these expressions, the transfer function can easily be computed for the electrical system model of amino acid chain of any arbitrary length.
    3. Results and discussions
    The genetic attributes are investigated by modeling sensor network for gene, which is tested on 40 gene databases (25 cancerous or hy-drophilic and 15 non-cancerous or hydrophobic) (Table 2 and Table 3) and the databases for the genes are downloaded from public domain (http://www.ncbi.nlm.nih.gov; http://cgap.nci.nih.gov; http://www. genecards.org). The electrical responses of the sensor are simulated in MATLAB (version R2009b) environment.
    3.1. Behavior analysis of sensor network using bode plot
    The sensor networks representing genes are analyzed in frequency domain by observing their spectrums. Differentiation between 
    Fig. 2. Gene sensor responses in phase for cancer and non-cancer genes. The phase response for cancer gene shows negative value whereas non-cancer gene shows positive value at higher frequency. A. NUP214 vs. LOC107815086 gene phase plots. B. spr1652 vs. SHBG gene phase plots.
    cancerous and non-cancerous genes is obtained by investigating the correlation of the gene features and their simulated system behavior.
    The sensor behavior is studied using Bode magnitude and phase
    Fig. 3. Confusion matrix of binary classifier for gene classification. Genes are classified based on their hydrophilicity and hydrophobicity features.
    Table 5 Performance evaluation metrics for genes at different frequency.
    Frequency (Hz) Accuracy MCC TP rate TN rate Precision (P) Precision (N)
    values in the frequency range of 1 Hz to 1 MHz as detailed in Table 4. There are no marked differences observed in amplitude values; hence only the phase values of all the electrical system models representing genes are considered. 477-30-5 The plots in Fig. 2 are obtained by cascading the amino acid circuit models, where each amino acid having constant Rb of 7 Ω for backbone circuit and LSC or CSC of different values depending on hydropathy index values of hydrophobic or hydrophilic amino acid for side chain. Fig. 2 exhibits significant differences in phase responses between cancerous and non-cancerous genes and the phase values are markedly distinguished from each other within the frequency range of 50 kHz to 1 MHz.
    The simulated results (Fig. 2) for cancerous genes show negative phase at higher frequency as they are modeled by cascaded RC parallel circuit, which indicates these genes contain large amount of hydrophilic or polar amino acids. Whereas non-cancerous genes exhibit positive phase at higher frequency since they realized by RL parallel circuit, indicates they made up of large amount of hydrophobic amino acids. Therefore the cancerous and non-cancerous genes exhibit polar and 
    nonpolar characteristics respectively, which are clearly observed by the corresponding simulated phase responses, and the sensor realization is truly matched with the biological features (Stranzl et al., 2012) of genes.
    3.2. Performance evaluation of sensor characteristics
    The gene datasets, collected from the national website for health-care, are classified by modeling sensor. The sensor performance is judged by receiver operating characteristic (ROC) curve and analyzed using the following measurement metrics:
    • Accuracy is the ratio of the number of correctly classified genes to the total number of genes.
    • True positive rate or sensitivity (TPR) is the ratio of the number of correctly classified genes from the positive class (TP) i.e. cancer to the number of all genes from the positive class (TP + FN).