Limits...
Novel use of matched filtering for synaptic event detection and extraction.

Shi Y, Nenadic Z, Xu X - PLoS ONE (2010)

Bottom Line: This new technique was applied to quantify and compare the EPSCs obtained from excitatory pyramidal cells and fast-spiking interneurons.In addition, our technique has been further applied to the detection and analysis of inhibitory postsynaptic current (IPSC) responses.Given the general purpose of our matched filtering and signal recognition algorithms, we expect that our technique can be appropriately modified and applied to detect and extract other types of electrophysiological and optical imaging signals.

View Article: PubMed Central - PubMed

Affiliation: Department of Anatomy and Neurobiology, School of Medicine, University of California Irvine, Irvine, California, USA.

ABSTRACT
Efficient and dependable methods for detection and measurement of synaptic events are important for studies of synaptic physiology and neuronal circuit connectivity. As the published methods with detection algorithms based upon amplitude thresholding and fixed or scaled template comparisons are of limited utility for detection of signals with variable amplitudes and superimposed events that have complex waveforms, previous techniques are not applicable for detection of evoked synaptic events in photostimulation and other similar experimental situations. Here we report on a novel technique that combines the design of a bank of approximate matched filters with the detection and estimation theory to automatically detect and extract photostimluation-evoked excitatory postsynaptic currents (EPSCs) from individually recorded neurons in cortical circuit mapping experiments. The sensitivity and specificity of the method were evaluated on both simulated and experimental data, with its performance comparable to that of visual event detection performed by human operators. This new technique was applied to quantify and compare the EPSCs obtained from excitatory pyramidal cells and fast-spiking interneurons. In addition, our technique has been further applied to the detection and analysis of inhibitory postsynaptic current (IPSC) responses. Given the general purpose of our matched filtering and signal recognition algorithms, we expect that our technique can be appropriately modified and applied to detect and extract other types of electrophysiological and optical imaging signals.

Show MeSH
Detection of EPSCs with matched filters.A. In the top portion, the original and high-pass filtered traces are aligned. The raw trace, which is high-pass filtered with a Butterworth filter, contains a large direct response and synaptically mediated responses. The vertical dashed line indicates the photostimulation laser onset. The direct response window is defined as 10 ms within the laser onset. The filtered data trace is convolved with all the filters (a total of 18 matched filters in this case) from the bank, and the convolution traces (one for each filter) are compared to a threshold. In the bottom portion of A are shown 6 example convolution traces (green) produced with 6 matched filters (red) and their original EPSC templates (black). The detection threshold (dashed) is chosen as μ−1.2σ (11 pA), where μ is the mean value (28.4 pA) of cmax obtained in the training stage from the bank of 18 filters and σ is its standard deviation (14.5 pA). All the samples of the convolution traces that cross the detection threshold form the supratheshold segments (red squares); each trace may has its own set of suprathreshold segments. The red crosses illustrate the centers of mass of the supratheshold segments and represent potential EPSC occurrence times, while the black crosses are determined as identified EPSC peaks. As the arrow heads point out, more than one EPSCs can be identified within one suprathreshold segment. As the convolution values of the direct response are large and exceed the outlier threshold, defined as μ+4 σ (86 pA) within W ms (i.e., 30 ms) after the laser onset, the direct response is not detected as an EPSC response. B is similarly formatted as A, and shows another example to detect both a direct response and synaptically mediated responses. The direct response in B is relatively small, and its peak values of the convolution traces do not exceed the outlier threshold. But the response is correctly identified as a direct response, because the leading edge of the response is located within the 10 ms direct response window.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2991367&req=5

pone-0015517-g002: Detection of EPSCs with matched filters.A. In the top portion, the original and high-pass filtered traces are aligned. The raw trace, which is high-pass filtered with a Butterworth filter, contains a large direct response and synaptically mediated responses. The vertical dashed line indicates the photostimulation laser onset. The direct response window is defined as 10 ms within the laser onset. The filtered data trace is convolved with all the filters (a total of 18 matched filters in this case) from the bank, and the convolution traces (one for each filter) are compared to a threshold. In the bottom portion of A are shown 6 example convolution traces (green) produced with 6 matched filters (red) and their original EPSC templates (black). The detection threshold (dashed) is chosen as μ−1.2σ (11 pA), where μ is the mean value (28.4 pA) of cmax obtained in the training stage from the bank of 18 filters and σ is its standard deviation (14.5 pA). All the samples of the convolution traces that cross the detection threshold form the supratheshold segments (red squares); each trace may has its own set of suprathreshold segments. The red crosses illustrate the centers of mass of the supratheshold segments and represent potential EPSC occurrence times, while the black crosses are determined as identified EPSC peaks. As the arrow heads point out, more than one EPSCs can be identified within one suprathreshold segment. As the convolution values of the direct response are large and exceed the outlier threshold, defined as μ+4 σ (86 pA) within W ms (i.e., 30 ms) after the laser onset, the direct response is not detected as an EPSC response. B is similarly formatted as A, and shows another example to detect both a direct response and synaptically mediated responses. The direct response in B is relatively small, and its peak values of the convolution traces do not exceed the outlier threshold. But the response is correctly identified as a direct response, because the leading edge of the response is located within the 10 ms direct response window.

Mentions: Our new technique of matched filtering can be effectively applied to detection of photostimulation-evoked EPSCs, as exemplified in Figure 2. The raw data trace was first high-pass filtered with a Butterworth filter, which reduces the effect of the direct response and low frequency drifts (see the Methods). The filtered data trace is then convolved with all the matched filters from the bank, with potential EPSCs having better fitting of the templates and exhibiting larger convolution amplitudes. The examples of matched filters and their convolution traces are shown in Figure 2A and B. Note that the filters have different shapes or waveforms, based upon a range of EPSC templates selected from experimental datasets. For each candidate EPSC, given that multiple samples of a convolution trace from one matched filter are likely to exceed the threshold, and considering that multiple convolution traces can exceed the threshold, the centers of mass of all the suprathreshold segments in all convolution traces are calculated. The arrival time of candidate EPSCs can be found in the vicinity of the center-of- mass points (see the Methods for details).


Novel use of matched filtering for synaptic event detection and extraction.

Shi Y, Nenadic Z, Xu X - PLoS ONE (2010)

Detection of EPSCs with matched filters.A. In the top portion, the original and high-pass filtered traces are aligned. The raw trace, which is high-pass filtered with a Butterworth filter, contains a large direct response and synaptically mediated responses. The vertical dashed line indicates the photostimulation laser onset. The direct response window is defined as 10 ms within the laser onset. The filtered data trace is convolved with all the filters (a total of 18 matched filters in this case) from the bank, and the convolution traces (one for each filter) are compared to a threshold. In the bottom portion of A are shown 6 example convolution traces (green) produced with 6 matched filters (red) and their original EPSC templates (black). The detection threshold (dashed) is chosen as μ−1.2σ (11 pA), where μ is the mean value (28.4 pA) of cmax obtained in the training stage from the bank of 18 filters and σ is its standard deviation (14.5 pA). All the samples of the convolution traces that cross the detection threshold form the supratheshold segments (red squares); each trace may has its own set of suprathreshold segments. The red crosses illustrate the centers of mass of the supratheshold segments and represent potential EPSC occurrence times, while the black crosses are determined as identified EPSC peaks. As the arrow heads point out, more than one EPSCs can be identified within one suprathreshold segment. As the convolution values of the direct response are large and exceed the outlier threshold, defined as μ+4 σ (86 pA) within W ms (i.e., 30 ms) after the laser onset, the direct response is not detected as an EPSC response. B is similarly formatted as A, and shows another example to detect both a direct response and synaptically mediated responses. The direct response in B is relatively small, and its peak values of the convolution traces do not exceed the outlier threshold. But the response is correctly identified as a direct response, because the leading edge of the response is located within the 10 ms direct response window.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2991367&req=5

pone-0015517-g002: Detection of EPSCs with matched filters.A. In the top portion, the original and high-pass filtered traces are aligned. The raw trace, which is high-pass filtered with a Butterworth filter, contains a large direct response and synaptically mediated responses. The vertical dashed line indicates the photostimulation laser onset. The direct response window is defined as 10 ms within the laser onset. The filtered data trace is convolved with all the filters (a total of 18 matched filters in this case) from the bank, and the convolution traces (one for each filter) are compared to a threshold. In the bottom portion of A are shown 6 example convolution traces (green) produced with 6 matched filters (red) and their original EPSC templates (black). The detection threshold (dashed) is chosen as μ−1.2σ (11 pA), where μ is the mean value (28.4 pA) of cmax obtained in the training stage from the bank of 18 filters and σ is its standard deviation (14.5 pA). All the samples of the convolution traces that cross the detection threshold form the supratheshold segments (red squares); each trace may has its own set of suprathreshold segments. The red crosses illustrate the centers of mass of the supratheshold segments and represent potential EPSC occurrence times, while the black crosses are determined as identified EPSC peaks. As the arrow heads point out, more than one EPSCs can be identified within one suprathreshold segment. As the convolution values of the direct response are large and exceed the outlier threshold, defined as μ+4 σ (86 pA) within W ms (i.e., 30 ms) after the laser onset, the direct response is not detected as an EPSC response. B is similarly formatted as A, and shows another example to detect both a direct response and synaptically mediated responses. The direct response in B is relatively small, and its peak values of the convolution traces do not exceed the outlier threshold. But the response is correctly identified as a direct response, because the leading edge of the response is located within the 10 ms direct response window.
Mentions: Our new technique of matched filtering can be effectively applied to detection of photostimulation-evoked EPSCs, as exemplified in Figure 2. The raw data trace was first high-pass filtered with a Butterworth filter, which reduces the effect of the direct response and low frequency drifts (see the Methods). The filtered data trace is then convolved with all the matched filters from the bank, with potential EPSCs having better fitting of the templates and exhibiting larger convolution amplitudes. The examples of matched filters and their convolution traces are shown in Figure 2A and B. Note that the filters have different shapes or waveforms, based upon a range of EPSC templates selected from experimental datasets. For each candidate EPSC, given that multiple samples of a convolution trace from one matched filter are likely to exceed the threshold, and considering that multiple convolution traces can exceed the threshold, the centers of mass of all the suprathreshold segments in all convolution traces are calculated. The arrival time of candidate EPSCs can be found in the vicinity of the center-of- mass points (see the Methods for details).

Bottom Line: This new technique was applied to quantify and compare the EPSCs obtained from excitatory pyramidal cells and fast-spiking interneurons.In addition, our technique has been further applied to the detection and analysis of inhibitory postsynaptic current (IPSC) responses.Given the general purpose of our matched filtering and signal recognition algorithms, we expect that our technique can be appropriately modified and applied to detect and extract other types of electrophysiological and optical imaging signals.

View Article: PubMed Central - PubMed

Affiliation: Department of Anatomy and Neurobiology, School of Medicine, University of California Irvine, Irvine, California, USA.

ABSTRACT
Efficient and dependable methods for detection and measurement of synaptic events are important for studies of synaptic physiology and neuronal circuit connectivity. As the published methods with detection algorithms based upon amplitude thresholding and fixed or scaled template comparisons are of limited utility for detection of signals with variable amplitudes and superimposed events that have complex waveforms, previous techniques are not applicable for detection of evoked synaptic events in photostimulation and other similar experimental situations. Here we report on a novel technique that combines the design of a bank of approximate matched filters with the detection and estimation theory to automatically detect and extract photostimluation-evoked excitatory postsynaptic currents (EPSCs) from individually recorded neurons in cortical circuit mapping experiments. The sensitivity and specificity of the method were evaluated on both simulated and experimental data, with its performance comparable to that of visual event detection performed by human operators. This new technique was applied to quantify and compare the EPSCs obtained from excitatory pyramidal cells and fast-spiking interneurons. In addition, our technique has been further applied to the detection and analysis of inhibitory postsynaptic current (IPSC) responses. Given the general purpose of our matched filtering and signal recognition algorithms, we expect that our technique can be appropriately modified and applied to detect and extract other types of electrophysiological and optical imaging signals.

Show MeSH