JOURNAL ARTICLE

Feature Selection Using Linear Support Vector Machines

Janez BrankMarko Grobelnik

Year: 2002 Journal:   The Journal of Physiology Vol: 491 ( Pt 3) Pages: 18-18   Publisher: Wiley

Abstract

1. ATP-evoked currents and Ca2+ signals were simultaneously recorded in isolated inner hair cells (IHC) of guinea-pig cochlea by combining conventional whole-cell or perforated patch clamp recording with indo-1 dual emission microfluorometry. 2. In most IHCs, voltage clamped near resting membrane potential (-40 mV), extracellular ATP evoked a rapid inward current (time constant, 150 ms). This current was concomitant with a slow rise in [Ca2+]i (time constant, 5 s). The ATP-evoked inward currents could be repeated several times with only a small run-down in amplitude (< 10%), while the ATP-evoked Ca2+ responses showed a rapid run-down (> 80% at the third ATP application). 3. The current-voltage relationship of ATP-evoked currents showed a reversal potential at -11 +/- 6 mV (n = 8), suggesting that ATP essentially activated a non-specific cationic conductance. On the contrary, the amplitude of the ATP-evoked Ca2+ responses did not show significant dependence on holding membrane potential. 4. The Ca2+ response showed an apparent Kd for ATP (EC50, 1.8 +/- 0.3 microM; Hill coefficient, 1.0 +/- 0.1) eightfold smaller than for the evoked currents (EC50, 13.7 +/- 3.0 microM; Hill coefficient, 2.0 +/- 0.7). 5. Perfusion with high extracellular Ca2+ solution (10 mM CaCl2) reduced the amplitude of the ATP-evoked currents by 90%, while perfusion with zero Ca2+ solution increased it by more than 100%. However, similar variations in external Ca2+ concentration did not change the amplitude of the ATP-evoked Ca2+ responses. Furthermore, intracellular heparin (1 mg mL-1), a potent inhibitor of InsP3 receptors, did not significantly change the amplitude of ATP-evoked currents but reduced the ATP-evoked Ca2+ response, suggesting again that the latter is related to Ca2+ release from intracellular stores. 6. The results suggested that two types of P2-purinergic receptor are expressed in IHCs: ATP-gated ion channels and ATP-activated metabotropic receptors. At submicromolar ATP concentrations, the metabotropic receptors raising intracellular [Ca2+] would hyperpolarize IHCs via Ca(2+)-sensitive K+ channels. The ATP-gated ion channels activated at higher ATP concentrations would mainly have a depolarizing effect on IHCs.

Keywords:
Feature selection Feature vector Support vector machine Pattern recognition (psychology) Artificial intelligence Computer science Linear classifier Classifier (UML) Feature (linguistics) Data mining

Metrics

75
Cited By
3.34
FWCI (Field Weighted Citation Impact)
17
Refs
0.93
Citation Normalized Percentile
Is in top 1%
Is in top 10%

Citation History

Topics

Text and Document Classification Technologies
Physical Sciences →  Computer Science →  Artificial Intelligence
Face and Expression Recognition
Physical Sciences →  Computer Science →  Computer Vision and Pattern Recognition
Image Retrieval and Classification Techniques
Physical Sciences →  Computer Science →  Computer Vision and Pattern Recognition

Related Documents

JOURNAL ARTICLE

Feature selection for linear support vector machines

Zhizheng LiangTuo Zhao

Year: 2006 Vol: 5 Pages: 606-609
JOURNAL ARTICLE

Feature Selection Using Support Vector Machines

Janez BrankMarko GrobelnikN. Milić-FraylingDunja Mladenić

Journal:   WIT transactions on information and communication technologies Year: 2002 Vol: 28
BOOK-CHAPTER

Linear Penalization Support Vector Machines for Feature Selection

Jaime MirandaRicardo MontoyaRichard W. Weber

Lecture notes in computer science Year: 2005 Pages: 188-192
JOURNAL ARTICLE

Feature Selection using Fuzzy Support Vector Machines

Hong XiaBao Qing Hu

Journal:   Fuzzy Optimization and Decision Making Year: 2006 Vol: 5 (2)Pages: 187-192
© 2026 ScienceGate Book Chapters — All rights reserved.