Dataset Information

In Silico Prediction of Gamma-Aminobutyric Acid Type-A Receptors Using Novel Machine-Learning-Based SVM and GBDT Approaches.

ABSTRACT: Gamma-aminobutyric acid type-A receptors (GABAARs) belong to multisubunit membrane spanning ligand-gated ion channels (LGICs) which act as the principal mediators of rapid inhibitory synaptic transmission in the human brain. Therefore, the category prediction of GABAARs just from the protein amino acid sequence would be very helpful for the recognition and research of novel receptors. Based on the proteins' physicochemical properties, amino acids composition and position, a GABAAR classifier was first constructed using a 188-dimensional (188D) algorithm at 90% cd-hit identity and compared with pseudo-amino acid composition (PseAAC) and ProtrWeb web-based algorithms for human GABAAR proteins. Then, four classifiers including gradient boosting decision tree (GBDT), random forest (RF), a library for support vector machine (libSVM), and k-nearest neighbor (k-NN) were compared on the dataset at cd-hit 40% low identity. This work obtained the highest correctly classified rate at 96.8% and the highest specificity at 99.29%. But the values of sensitivity, accuracy, and Matthew's correlation coefficient were a little lower than those of PseAAC and ProtrWeb; GBDT and libSVM can make a little better performance than RF and k-NN at the second dataset. In conclusion, a GABAAR classifier was successfully constructed using only the protein sequence information.

SUBMITTER: Liao Z

PROVIDER: S-EPMC4992803 | biostudies-literature | 2016

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

In Silico Prediction of Gamma-Aminobutyric Acid Type-A Receptors Using Novel Machine-Learning-Based SVM and GBDT Approaches.

Liao Zhijun Z Huang Yong Y Yue Xiaodong X Lu Huijuan H Xuan Ping P Ju Ying Y

BioMed research international 20160808

Gamma-aminobutyric acid type-A receptors (GABAARs) belong to multisubunit membrane spanning ligand-gated ion channels (LGICs) which act as the principal mediators of rapid inhibitory synaptic transmission in the human brain. Therefore, the category prediction of GABAARs just from the protein amino acid sequence would be very helpful for the recognition and research of novel receptors. Based on the proteins' physicochemical properties, amino acids composition and position, a GABAAR classifier was ...[more]

PMID: 27579307

Similar Datasets

Project description:γ-aminobutyric acid type A receptors (GABAARs), the major inhibitory neurotransmitter receptors in the mammalian central nervous system, are arguably the most challenging member of the pentameric Cys-loop receptors to study due to their heteromeric structure. When two or more subunits are expressed together in heterologous systems, receptors of variable subunit type, ratio, and orientation can form, precluding accurate interpretation of data from functional studies. Subunit concatenation is a technique that involves the linking of individual subunits and in theory allows the precise control of the uniformity of expressed receptors. In reality, the resulting concatemers from widely used constructs are flexible in their orientation and may therefore assemble with themselves or free GABAAR subunits in unexpected ways. In this study, we examine functional responses of receptors from existing concatenated constructs and describe refinements necessary to allow expression of uniform receptor populations. We find that dimers from two commonly used concatenated constructs, β-23-α and α-10-β, assemble readily in both the clockwise and the counterclockwise orientations when coexpressed with free subunits. Furthermore, we show that concatemers formed from new tetrameric α-10-β-α-β and α-10-β-α-γ constructs also assemble in both orientations with free subunits to give canonical αβγ receptors. To restrict linker flexibility, we systematically shorten linker lengths of dimeric and pentameric constructs and find optimized constructs that direct the assembly of GABAARs only in one orientation, thus eliminating the ambiguity associated with previously described concatemers. Based on our data, we revisit some noncanonical GABAAR configurations proposed in recent years and explain how the use of some concatenated constructs may have led to wrong conclusions. Our results help clarify current contradictions in the literature regarding GABAAR subunit stoichiometry and arrangement. The lessons learned from this study may guide future efforts in understanding other related heteromeric receptors.

Project description:ObjectiveHuman hypothalamic hamartomas (HHs) are highly associated with treatment-resistant gelastic seizures. HHs are intrinsically epileptogenic, although the basic cellular mechanisms responsible for seizure activity are unknown. Altered gamma-aminobutyric acid (GABA) function can contribute to epileptogenesis in humans and animal models. Recently, functional GABA(A) receptor (GABA(A) R) rundown has been described in surgically resected human temporal lobe epilepsy tissue. We asked whether functional GABA(A) R rundown also occurs in human HH neurons.MethodsGABA(A) R-mediated currents were measured using perforated patch-clamp recordings in single neurons acutely dissociated from surgically resected HH tissue. In addition, functional GABA(A) Rs were expressed in Xenopus oocytes after microinjection with membrane fractions from either HH or control hypothalamus, and were studied with 2-electrode voltage-clamp recordings.ResultsPerforated patch-clamp recordings in dissociated HH neurons showed that repetitive exposure to GABA (5 consecutive exposures to 0.1 mM GABA with 1-second duration and at 20-second intervals) induced a time-dependent rundown of whole-cell currents in small HH neurons, whereas large HH neurons showed much less rundown using the same protocol. Functional rundown was not observed in HH neurons with repetitive exposure to glycine or glutamate. Two-electrode voltage-clamp recordings (6 consecutive exposures to 1 mM GABA with 10-second duration and at 40-second intervals) induced GABA current rundown in Xenopus oocytes microinjected with HH membrane proteins, but not in the oocytes expressing hypothalamic membrane proteins derived from human autopsy controls. Functional rundown of GABA currents was significantly attenuated by intracellular application of adenosine triphosphate or the nonspecific phosphatase inhibitor, okadaic acid.InterpretationNeurons from surgically resected human HH demonstrate functional rundown of GABA(A) R-mediated transmembrane currents in response to GABA agonist exposure. Rundown may be a marker for impaired GABAergic function and a contributing mechanism for seizure genesis within HH tissue.

Dataset Information

In Silico Prediction of Gamma-Aminobutyric Acid Type-A Receptors Using Novel Machine-Learning-Based SVM and GBDT Approaches.

Publications

In Silico Prediction of Gamma-Aminobutyric Acid Type-A Receptors Using Novel Machine-Learning-Based SVM and GBDT Approaches.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets