Estimation method, estimation device, estimation system, and program

Abstract

This estimation method is performed by an estimation device equipped with a control unit which executes an estimation processing that is a processing for estimating the position of a synaptic connection of a nerve cell that is a target to be analyzed and is in contact with a plurality of electrodes. In the estimation processing, the estimation is performed on the basis of: electrode position information which indicates the positions of the electrodes; neurite position information which indicates the positions of neurites extending from the nerve cell; and cue information which indicates an estimation signal pair, wherein the estimation signal pair is included in estimation signal pairs each of which is a pair of ignition electrodes that are electrodes in which a propagated electrical signal has a spike-like waveform, and the estimation signal pair satisfies at least such an extraction requirement that, with respect to firing that is a phenomenon where electrical signals each having a spike-like waveform are generated, the difference between the time at which the firing occurs in one of the pair of the ignition electrodes and the time at which the ignition occurs in the other of the pair of the ignition electrodes is longer than the time required for the electrical signal to travel a straight-line distance between the above-mentioned one of the pair of the ignition electrodes and the above-mentioned other of the pair of the ignition electrodes at a speed at which the electrical signal propagates through an axon.

Description

Estimation method, estimation device, estimation system, and program

[0001] This application claims priority to Japanese Patent Application No. 2024-044091, filed on March 19, 2024, the contents of which are incorporated herein by reference.

[0002] Research into neurons is active due to their potential use in drug discovery, etc. However, estimating the location of synaptic connections or the pairs of cells that make synaptic connections can be difficult and burdensome.

[0003] The present invention aims to provide an estimation method, an estimation device, an estimation system, and a program that can reduce the burden required to estimate the location of synaptic connections or the pair of cells that are synaptically connected.

[0004] One aspect of the present invention includes a control unit that executes an estimation process that estimates the positions of synaptic connections of neurons to be analyzed that are in contact with a plurality of electrodes, or the pair of cells that are synaptically connected, and the estimation process estimates the synaptic connections or the positions of the pair of cells that are synaptically connected based on electrode position information that indicates the positions of the electrodes, neurite position information that indicates the positions of neurites extending from the neurons, and clue information that indicates an estimation signal pair that satisfies at least an extraction condition that, among estimation signal pairs that are pairs of firing electrodes that are electrodes that are electrodes that propagate a spike-shaped waveform, the difference between the timing at which firing occurs in one firing electrode of the pair and the timing at which firing occurs in the other firing electrode of the pair for firing, which is the generation of the spike-shaped waveform electrical signal, is longer than the time it takes for an electrical signal to travel a straight line distance between the one and the other at a standard axonal propagation velocity that is the velocity at which the electrical signal propagates through an axon. The estimation method is executed by an estimation device, and includes a control step of executing an estimation process that estimates the positions of synaptic connections of neurons to be analyzed that contact multiple electrodes, or pairs of cells that are synaptically connected.

[0005] According to the present invention, it is possible to reduce the burden required for estimating the location of synaptic connections or the pair of cells that are synaptically connected.

[0006] 1 is an explanatory diagram illustrating an estimation system according to an embodiment. A first explanatory diagram illustrating an example of an experimental result according to an embodiment. A second explanatory diagram illustrating an example of an experimental result according to an embodiment. A third explanatory diagram illustrating an example of an experimental result according to an embodiment. A fourth explanatory diagram illustrating an example of an experimental result according to an embodiment. A fifth explanatory diagram illustrating an example of an experimental result according to an embodiment. A sixth explanatory diagram illustrating an example of an experimental result according to an embodiment. A seventh explanatory diagram illustrating an example of an experimental result according to an embodiment. An eighth explanatory diagram illustrating an example of an experimental result according to an embodiment. A ninth explanatory diagram illustrating an example of an experimental result according to an embodiment. A tenth explanatory diagram illustrating an example of an experimental result according to an embodiment. An eleventh explanatory diagram illustrating an example of an experimental result according to an embodiment. A twelfth explanatory diagram illustrating an example of an experimental result according to an embodiment. A diagram illustrating an example of the hardware configuration of an estimation device according to an embodiment. A flowchart illustrating an example of a processing flow executed by an estimation device according to an embodiment.

[0007] 1 is an explanatory diagram illustrating an estimation system 100 according to an embodiment. The estimation system 100 estimates the positions of synaptic connections in a neuron to be analyzed, or pairs of cells that are synaptically connected. The estimation system 100 includes a signal detector 1 and an estimation device 2. The signal detector 1 includes a plurality of electrodes 11 that are in contact with the neuron to be analyzed.

[0008] Since the object of analysis is a neuron, it has a chemical synapse, which is a gap of several tens of nanometers formed between the axon of a presynaptic cell and the tip of the dendrite of a postsynaptic cell. In a neuron, an electrical signal propagates through this gap due to the propagation of chemical substances. Specifically, this electrical signal is a spike-shaped electrical signal (hereinafter referred to as a "spike signal"). Therefore, when a spike signal occurs at the part of the neuron in contact with the electrode 11, this spike signal is generated. Hereinafter, the part of the neuron in contact with the electrode 11 is referred to as the contact point of the electrode 11. Hereinafter, the generation of a spike signal is referred to as firing, the fired electrode 11 is referred to as the firing electrode, and the timing at which firing occurs is referred to as the firing timing. When the electrode 11 fires, the signal detector 1 detects the spike signal generated in the object of analysis.

[0009] The signal detector 1 is, for example, a microelectrode array (MEA). Note that the signal detector 1 is not limited to a microelectrode array as long as it can measure electrical signals at a plurality of locations.

[0010] The estimation device 2 includes a control unit 21 including a processor 91 such as a central processing unit (CPU), a graphics processing unit (GPU), or a neural network processing unit (NPU), and a memory 92 connected via a bus.

[0011] The control unit 21 executes an estimation process. The estimation process is a process for estimating the position of a synaptic connection of an object of analysis or a pair of synaptically connected cells based on electrode position information, neurite position information, and clue information. The electrode position information is information indicating the position of each electrode 11. The neurite position information is information indicating the position of a neurite extending from an object of analysis neuron. Note that the neurite may be, for example, a nerve axon. Since the neurite position information indicates the position of a neurite extending from an object of analysis neuron and the electrode position information indicates the position of the electrode 11, information on the contact point of each electrode 11 can be obtained from the electrode position information and the neurite position information. Furthermore, information indicating which electrode 11 the position of a neurite of interest contacts can also be obtained from the electrode position information and the neurite position information. Note that the position of the electrode 11 indicated by the electrode position information is, for example, the position of each electrode 11 on the signal detector 1. In such a case, the position indicated by the neurite position information is the position of a neurite extending from an object of analysis neuron on the signal detector 1. The position of the electrode 11 indicated by the electrode position information is the position of each electrode 11 in a specified space, such as a laboratory, and in such a case, the position indicated by the neurite position information is the position of the neurite extending from the neuron being analyzed in the above space.

[0012] The clue information is information indicating estimation signal pairs, which are pairs of firing electrodes, that at least satisfy an extraction condition, where the difference in firing timing between a first electrode, which is one of the pair of firing electrodes, and a second electrode, which is the other of the pair of firing electrodes, is longer than the time it takes for an electrical signal to travel a linear distance between the first electrode and the second electrode at a standard axon propagation velocity (hereinafter referred to as the “standard time”).

[0013] The estimation signal pair indicated by the clue information may further satisfy the condition that, for example, the first electrode is a trigger electrode and the second electrode is a reaction electrode. The trigger electrode is an electrode that fires earlier than the other of the two firing electrodes in the estimation signal pair. The reaction electrode is an electrode that is not a trigger electrode in the two firing electrodes in the estimation signal pair. Therefore, the reaction electrode is an electrode that fires later than the other of the two firing electrodes in the estimation signal pair.

[0014] The trigger electrode may be located immediately below or around the neurite.

[0015] <Effects of the Estimation Process> The propagation speed of electrical signals at synaptic connections is slower than the propagation speed of electrical signals through axons. Therefore, it is estimated that a synaptic connection exists between two electrodes of an estimation signal pair (hereinafter referred to as a "matching estimation signal pair") that meets the extraction conditions. The clue information merely indicates the matching estimation signal pair. Therefore, the estimation process uses electrode position information and neurite position information to estimate where on the neurite the position between each electrode of the estimation signal pair corresponds. In this way, the position of the synaptic connection on the neurite is estimated. Since the location of the synaptic connection can also be estimated as the pair of cells that are synaptically connected, estimating the location of the synaptic connection can also be considered as estimating the pair of cells that are synaptically connected. Therefore, the estimation process can reduce the burden required to estimate the location of the synaptic connection or the pair of cells that are synaptically connected.

[0016] In the estimation process, for example, the position of the synaptic connection or the pair of cells that are synaptically connected may be estimated based on electrode position information, neurite position information, and clue information by comparing the position of each electrode of the estimation signal pair indicated by the electrode position information and the position indicated by the clue information with the position of the neurite indicated by the neurite position information.

[0017] The clue information used by the control unit 21 in the estimation process is, for example, calculated and acquired by the control unit 21 itself based on the signal detection results by the signal detector 1. However, the clue information does not have to be calculated and acquired by the control unit 21 itself. For example, the clue information may be calculated and acquired by another control unit, such as a control unit of another device different from the estimation device 2. In such a case, the control unit 21 may acquire the clue information thus acquired, for example, via communication, and use it in the estimation process. The clue information may also be acquired by a person through manual calculation. In such a case, the control unit 21 may acquire the clue information obtained by manual calculation through input by the user, and use it in the estimation process.

[0018] <Explanation of clue information using experimental results> The clue information will be explained using examples of experimental results and figures. The experiment was started in a state where time series data indicating the amplitude of electrical signals, including spike-like signals, had been obtained for all electrodes 11 provided in the signal detector 1. In the experiment, the time interval between firings (inter-spike interval) was calculated using this time history data.

[0019] In the analysis of the experiment, a portion of 200 seconds of data was used to shorten the calculation time. In the experiment, the firing time interval between each electrode 11 and the surrounding electrodes was calculated, and locations where the same delay time occurred multiple times were extracted. In the experiment, locations where the same delay time occurred three or more times were extracted.

[0020] For simplicity of explanation, the following describes the results of an experiment on a portion of the analysis target (hereinafter referred to as the "explanation target portion"). That is, for simplicity of explanation, the results are based on the electrical signals generated in a portion of the electrodes 11 provided in the signal detector 1.

[0021] Fig. 2 is a first explanatory diagram illustrating an example of an experimental result in an embodiment. More specifically, Fig. 2 is a diagram illustrating an explanation target region used in the experiment. Fig. 2 shows the results of visualizing cell locations and axons. The region surrounded by box A1 in Fig. 2 is the explanation target region used in the experiment.

[0022] In the experiment, the explanation target region was set by the user. However, this does not necessarily have to be set by the user. For example, the analysis target region may be set by a device other than the estimation device 2 or the control unit 21 executing the analysis target region setting process. The analysis target region setting process is a process of setting the explanation target region according to a predetermined rule. The predetermined rule may be, for example, a method of automatically identifying a region where multiple high firing frequencies are concentrated. Specifically, the rule may divide a predetermined range into specified regions, calculate the number of high firing frequencies per area in each divided region, and prioritize extraction of regions with a predetermined number or more as explanation target regions. Alternatively, as clue information, the process may start when time series data indicating amplitudes of electrical signals including spike-like signals measured simultaneously at some electrodes 11 has been obtained. Note that firing frequency refers to the number of firings per unit time.

[0023] FIG. 3 is a second explanatory diagram illustrating an example of experimental results in an embodiment. FIG. 3 shows the positions of the firing electrodes in the signal detector 1. The positions of the firing electrodes are indicated by an x-y coordinate system. In FIG. 3, trigger electrodes are indicated by crosses and reaction electrodes are indicated by circles. Note that FIG. 3 includes multiple crosses and circles, but which crosses and which circles belong to the same estimation signal pair can be confirmed by looking at the diagram of the reaction electrodes for each trigger electrode shown in FIG. 6 (described later). FIG. 3 shows the results of multiple trigger electrodes superimposed on each other to facilitate understanding of the overview of the embodiment. This diagram is shown after the results shown in FIG. 6 (described later) have been obtained for ease of understanding. In FIG. 3, the time interval between the firing timing of the trigger electrode and the firing timing of the reaction electrode (hereinafter referred to as the "firing propagation time") for multiple estimation signal pairs is indicated by shading. In FIG. 3, "delay [ms]" indicates the firing propagation time in milliseconds. The firing propagation time is an example of the difference between the firing timing at which firing occurs at the first electrode and the firing timing at which firing occurs at the second electrode.

[0024] In Figure 3, the estimation signal pairs are displayed by distinguishing between the trigger electrode and the reaction electrode. However, several pieces of information are required for this display. First, electrode position information is required. Second, information indicating the estimation signal pairs (hereinafter referred to as "estimation signal pair presentation information") is required. Furthermore, information indicating which of the two electrodes is the trigger electrode and which is the reaction electrode for each estimation signal pair (hereinafter referred to as "front-rear information") is required.

[0025] The estimation signal pair presentation information and the front-back information may be obtained by a person based on electrical signals propagated to each electrode 11 provided in the signal detector 1, or may be obtained by processing by a device other than the estimation device 2 or the control unit 21. When the estimation signal pair presentation information is obtained by processing by a device other than the estimation device 2 or the control unit 21, the device other than the estimation device 2 or the control unit 21 obtains the estimation signal pair presentation information by executing estimation signal pair estimation processing. When the front-back information is obtained by processing by a device other than the estimation device 2 or the control unit 21, the device other than the estimation device 2 or the control unit 21 obtains the front-back information by executing front-back electrode estimation processing.

[0026] The estimate signal pair estimation process is a process that determines whether or not a fire has occurred for both of any two electrodes 11 in a pair based on the observation result information, and estimates the pair of electrodes 11 for which it has been determined that a fire has occurred for both electrodes as an estimate signal pair. Therefore, estimate signal pair presentation information is obtained by the estimate signal pair estimation process.

[0027] The observation result information is information indicating a time series, and is information indicating the time series of the amplitude of an electrical signal including a spike signal for each electrode 11 provided in the signal detector 1. Therefore, the processing based on the observation result information is processing based on the electrical signal propagated to each electrode 11 provided in the signal detector 1. Therefore, the observation result information includes information on which electrode 11 firing occurred and the timing of firing at that electrode 11.

[0028] The front-rear electrode estimation process is a process in which, based on the estimation signal pair presentation information and the observation result information, it is determined which of the two electrodes 11 has an earlier firing timing for each estimation signal pair, and the electrode whose firing timing is determined to be earlier than the other electrode is estimated to be the trigger electrode, and the electrode that is not the trigger electrode is estimated to be the reaction electrode. Therefore, front-rear information is obtained by the front-rear electrode estimation process.

[0029] Firing propagation times are also shown in Fig. 3. Therefore, information indicating firing propagation times for each estimation signal pair (hereinafter referred to as "firing propagation time information") is required for the display in Fig. 3.

[0030] The ignition propagation time information may be obtained manually based on the before-and-after information and the observation result information, or may be calculated by processing by a device other than the estimation device 2 or the control unit 21. When the ignition propagation time information is obtained by processing by a device other than the estimation device 2 or the control unit 21, the device other than the estimation device 2 or the control unit 21 executes an ignition propagation time acquisition process to obtain the ignition propagation time information. The ignition propagation time acquisition process is a process for obtaining a time interval obtained by subtracting the ignition timing of the trigger electrode from the ignition timing of the reaction electrode for each estimation signal pair, based on the before-and-after information and the observation result information. Therefore, the ignition propagation time information is obtained by the ignition propagation time acquisition process.

[0031] In the experiment, a synapse index was then calculated for each estimation signal pair shown in FIG. 3 (and therefore for each estimation signal pair indicated by the estimation signal pair presentation information, the before-and-after information, or the firing propagation time information). The synapse index is an index indicating whether the firing propagation time is longer than the standard time. Therefore, the synapse index is, for example, a value obtained by dividing the firing propagation time by the standard time. In the experiment, the synapse index was a value obtained by dividing the firing propagation time by the standard time.

[0032] Since the synaptic index indicates whether the firing propagation time is longer than the standard time, it is possible to determine whether each signal pair for estimation is a suitable signal pair for estimation from the value of each synaptic index for each signal pair for estimation. A signal pair for estimation whose synaptic index indicates that the firing propagation time is longer than the standard time is a suitable signal pair for estimation.

[0033] 4 is a third explanatory diagram illustrating an example of experimental results in an embodiment. More specifically, FIG. 4 is a diagram illustrating an example of the results of calculating synaptic indices for multiple estimation signal pairs. A estimation signal pair with a high synaptic indices indicates that excessive delay occurs in the electrical signal propagation from the trigger electrode of that estimation signal pair to the reaction electrode compared to when the propagation is assumed to be axonal. Therefore, a high synaptic index suggests that there is a high possibility that a reaction is occurring with a synaptic connection.

[0034] In the example of Fig. 4, the cross marks indicate the positions of the trigger electrodes and the circles indicate the positions of the reaction electrodes. In the example of Fig. 4, the results are shown for estimation signal pairs with firing propagation times of 5 ms or less and synaptic indices in the range of 4 to 5.

[0035] Electrode position information is required for the display in FIG. 4 . Furthermore, information indicating synaptic indices for each estimation signal pair (hereinafter referred to as "synaptic index information") is required. The synaptic index information may be obtained manually based on the electrode position information and firing propagation time information, or may be obtained by processing by a device other than the estimation device 2 or the control unit 21. When the synaptic index information is obtained by processing by a device other than the estimation device 2 or the control unit 21, the device other than the estimation device 2 or the control unit 21 obtains the synaptic index information by executing synaptic index estimation processing.

[0036] A synaptic index is an index that indicates whether a firing propagation time is longer than a standard time. Therefore, to obtain a synaptic index value, at least information indicating the electrode position and firing propagation time information are required. Therefore, the synaptic index estimation process is a process that calculates a synaptic index based on the electrode position information and firing propagation time information. The synaptic index information is obtained by the synaptic index estimation process.

[0037] An estimation signal pair whose firing propagation time is indicated as being longer than the standard time by the synaptic index information is a matching estimation signal pair. Therefore, a person, another device other than the estimation device 2, or the control unit 21 can obtain clue information based on the synaptic index information. When clue information is obtained from the synaptic index information through processing by the other device other than the estimation device 2 or the control unit 21, the other device other than the estimation device 2 or the control unit 21 obtains the clue information by performing filtering. The filtering is a process for obtaining, based on the synaptic index information, information indicating an estimation signal pair whose firing propagation time is indicated as being longer than the standard time.

[0038] In the experiment, further analysis was performed on the analysis target. For the further analysis, a matching estimation signal pair and an estimation signal pair (hereinafter referred to as a "first-kindred relative estimation signal pair") in which the reaction electrode of the matching estimation signal pair (hereinafter referred to as a "target electrode") was used as the reaction electrode, based on the display in Figure 4.

[0039] Fig. 5 is a fourth explanatory diagram illustrating an example of an experimental result in an embodiment. More specifically, Fig. 5 is a diagram illustrating an example of a target electrode used in the experiment. The electrode enclosed by box A2, which is located at an X coordinate of 3430 and a Y coordinate of 858, is the target electrode used in the experiment. Note that in the experiment, the inferential signal pair enclosed by box A3 was used as the first-kind relative inferential signal pair.

[0040] FIG. 6 is a fifth explanatory diagram illustrating an example of an experimental result in an embodiment. More specifically, it illustrates an example of a first-kind kinship inference signal pair. The firing electrode indicated by symbol A4 is the target electrode in the experiment. The firing electrode indicated by symbol A5 is the trigger electrode (hereinafter referred to as the "origin electrode") of the first-kind kinship inference signal pair whose firing propagation time satisfies a predetermined origin setting condition, which is a predetermined condition. In FIG. 6, the origin electrode is indicated by a cross, and other firing electrodes are indicated by circles. Note that the origin setting condition is, for example, a condition that the firing propagation time is longer than that of other first-kind kinship inference signal pairs. Note that the origin setting condition may also be a sufficiently long firing propagation time, for example, a condition that the firing propagation time is less than 24 hours.

[0041] FIG. 6 also shows the firing propagation time for each second-kind relative inferring signal pair by varying the shading of the symbol indicating its corresponding reaction electrode. The second-kind relative inferring signal pair is an inferring signal pair in which the origin electrode is used as the trigger electrode and the firing electrodes of the first-kind relative inferring signal pair, excluding the origin electrode, are used as reaction electrodes. In FIG. 6, "delay [ms]" indicates the firing propagation time in milliseconds. FIG. 6 shows that firing electrodes farther from the origin electrode tend to have longer firing propagation times.

[0042] The results in Figure 6 also indicate that the contact point of the firing electrode enclosed in box A6 is likely to be the axon of a pre-post-synaptic cell, and the contact point of the firing electrode enclosed in box A7 is likely to be a postsynaptic cell other than the pre-post-synaptic cell.

[0043] FIG. 7 is a sixth explanatory diagram illustrating an example of experimental results in an embodiment. FIG. 7 is an example of neurite position information, which is neurite position information for an analysis target in an experiment. In FIG. 7, boxes A6 and A7 in FIG. 6 are superimposed to show the correspondence with each electrode in FIG. 6. FIG. 7 shows that axons extend from the upper left to the lower right in the area surrounded by box A6. FIG. 7 shows that in the area surrounded by box A7, there is no axonal connection from the area surrounded by box A6, and other cells are present. Therefore, FIG. 7 indicates that the results in FIG. 6 are valid.

[0044] In the experiment, the relationship between the firing propagation time and the firing frequency, which is the number of firings, was further analyzed.

[0045] 8 is a seventh explanatory diagram illustrating an example of an experimental result in the embodiment. Fig. 8 shows the firing propagation time and the number of firings per unit time as an example of a portion of a signal pair for second-kind kinship estimation. Fig. 8 shows that a spike signal propagates from the origin electrode to the target electrode.

[0046] 9 is an eighth explanatory diagram illustrating an example of experimental results in an embodiment. The horizontal axis of FIG. 9 represents the linear distance from the origin electrode to the reaction electrode of the second-kind kinship inferring signal pair. The vertical axis of FIG. 9 represents the ignition propagation time. Data enclosed by box A8 represents data of the second-kind kinship inferring signal pair using the ignition electrode enclosed by box A6 in FIG. 6 as the reaction electrode. Data enclosed by box A9 represents data of the second-kind kinship inferring signal pair using the ignition electrode enclosed by box A7 in FIG. 6 as the reaction electrode.

[0047] Figure 9 shows that in the axon contacted by the firing electrode, enclosed by box A6, a spike signal propagates approximately 250 μm in 0.5 ms. This indicates that the propagation velocity is approximately 0.5 m / s. A propagation velocity of 0.5 m / s is the commonly known propagation velocity of spike signals in axons without myelin sheaths. Also, 0.5 m / s is the standard axonal propagation velocity.

[0048] Figure 9 shows that there is a significant delay in the propagation of spike signals to the contact points of the firing electrodes enclosed in box A7. Therefore, Figure 9 indicates that the propagation of spike signals to the contact points of the firing electrodes enclosed in box A7 is likely to be via synaptic connections rather than axonal propagation. Therefore, Figure 9 indicates that the contact points of the firing electrodes enclosed in box A6 are likely to be presynaptic cells, and the contact points of the firing electrodes enclosed in box A7 are likely to be postsynaptic cells.

[0049] FIG. 10 is a ninth explanatory diagram illustrating an example of experimental results in an embodiment. The horizontal axis of FIG. 10 indicates the linear distance from the origin electrode to the reaction electrode of the second-kind kinship inferring signal pair. The vertical axis of FIG. 10 indicates the number of firings. Data surrounded by box A10 indicates data on the second-kind kinship inferring signal pair using the firing electrode surrounded by box A6 in FIG. 6 as the reaction electrode. Data surrounded by box A11 indicates data on the second-kind kinship inferring signal pair using the firing electrode surrounded by box A7 in FIG. 6 as the reaction electrode.

[0050] Figure 10 shows that, in the axon at the contact point of the firing electrode enclosed in box A6, more than 90% of firings propagate downstream compared to the number of firings in the upstream portion of the axon. Figure 10 also shows that firings hardly propagate to the contact point of the firing electrode enclosed in box A7. Figure 10 also shows that there is a high possibility of a synaptic connection between the contact points of the firing electrodes enclosed in box A6 and A7. This is because, in response to firing of presynaptic cells, firing of postsynaptic cells does not occur every time, but only when certain conditions are met.

[0051] According to the results in Figure 10, there were approximately 30 firings on the presynaptic side and only 3 firings on the postsynaptic side, so the propagation strength of the synaptic connection between the presynaptic cell and the postsynaptic cell (hereinafter referred to as "first-type propagation strength") was calculated to be approximately 1 / 10.

[0052] In this way, the first-type propagation strength is obtained from the ratio of the number of firings of presynaptic cells to the number of firings of postsynaptic cells. The process of obtaining the first-type propagation strength (hereinafter referred to as the “first-type propagation strength obtaining process”) may be performed by a person, by a device other than the estimation device 2, or by the control unit 21.

[0053] Specifically, the first-type propagation strength acquisition process is a process that acquires, as the first-type propagation strength, the ratio expressed as the average number of firings per unit time of the postsynaptic side estimation signal pair to the average number of firings per unit time of the presynaptic side estimation signal pair, based on information indicating the presynaptic side estimation signal pair (hereinafter referred to as "presynaptic side estimation signal pair information"), information indicating the postsynaptic side estimation signal pair (hereinafter referred to as "postsynaptic side estimation signal pair information"), and information indicating the number of firings per unit time of each second-type relative estimation signal pair (hereinafter referred to as "number of firings information").

[0054] The presynaptic side inferential signal pair is a second-type relative inferential signal pair whose reaction electrode contacts a presynaptic cell, and the postsynaptic side inferential signal pair is a second-type relative inferential signal pair whose reaction electrode contacts a postsynaptic cell.

[0055] The presynaptic side estimation signal pair information may be prepared in advance by a person based on the neurite position information, estimation signal pair presentation information, and electrode position information, and provided to the control unit 21 via a predetermined interface. The postsynaptic side estimation signal pair information may be prepared in advance by a person based on the neurite position information, estimation signal pair presentation information, and electrode position information, and provided to the control unit 21 via a predetermined interface. The firing count information may be prepared in advance by a person based on the neurite position information, estimation signal pair presentation information, and electrode position information, and provided to the control unit 21 via a predetermined interface. However, the presynaptic side estimation signal pair information, the postsynaptic side estimation signal pair information, and the firing count information do not necessarily need to be prepared by a person, and may be obtained, for example, by the control unit 21 executing the following first information acquisition process.

[0056] The first information acquisition process includes a target electrode determination process, a second-kind relative inference signal pair determination process, a firing count estimation process, a postsynaptic side inference signal pair determination process, and a presynaptic side inference signal pair determination process. The target electrode determination process is a process for determining a target electrode based on the execution result of the filter process. The second-kind relative inference signal pair determination process is a process for determining a second-kind relative inference signal pair based on the result of the target electrode determination process and inference signal pair presentation information.

[0057] The firing count estimation process is a process for estimating the firing count for each second-kind relative inferring signal pair based on the results of the second-kind relative inferring signal pair determination process and the observation result information. Therefore, firing count information is obtained by the firing count estimation process.

[0058] The postsynaptic side inference signal pair determination process is a process for determining, as a postsynaptic side inference signal pair, a second-kind relative inference signal pair whose reaction electrode is located within a predetermined distance from the target electrode based on the result of the second-kind relative inference signal pair determination process and electrode position information. Therefore, the postsynaptic side inference signal pair determination process obtains postsynaptic side inference signal pair information.

[0059] The presynaptic side inference signal pair determination process is a process for determining, based on the result of the second-kind relative inference signal pair determination process and the electrode position information, a second-kind relative inference signal pair whose reaction electrode is located at a position farther than a predetermined distance from the target electrode as a presynaptic side inference signal pair. Therefore, the presynaptic side inference signal pair determination process obtains presynaptic side inference signal pair information.

[0060] Although the example of the propagation strength of the synaptic connection between a presynaptic cell and a postsynaptic cell has been described, the propagation strength of the synaptic connection between the axon of a presynaptic cell and a postsynaptic cell (hereinafter referred to as "second-type propagation strength") may also be obtained. The process of obtaining the second-type propagation strength (hereinafter referred to as "second-type propagation strength acquisition process") may be performed by a person, or by another device other than the estimation device 2 or the control unit 21.

[0061] Specifically, the second-type propagation strength acquisition process is a process for acquiring, as the second-type propagation strength, a ratio expressed as the average number of firings per unit time of the postsynaptic side estimation signal pair to the average number of firings per unit time of the presynaptic side estimation signal pair, based on the presynaptic side estimation signal pair information, the postsynaptic side estimation signal pair information, and the firing count information. The presynaptic side estimation signal pair is a second-type relative estimation signal pair whose reaction electrode contacts the neural axon of a presynaptic cell.

[0062] The presynaptic side estimation signal pair may be prepared in advance by a person and provided to the control unit 21 via a predetermined interface, for example. The postsynaptic side estimation signal pair information may be prepared in advance by a person and provided to the control unit 21 via a predetermined interface, for example. The firing count information may be prepared in advance by a person and provided to the control unit 21 via a predetermined interface, for example. However, the presynaptic side estimation signal pair, the postsynaptic side estimation signal pair information, and the firing count information do not necessarily need to be prepared by a person, and may be obtained by the control unit 21 executing the following second information acquisition process, for example, based on observation result information.

[0063] The second information acquisition process includes a target electrode determination process, a second-kind relative inference signal pair determination process, a firing count estimation process, and a presynaptic side inference signal pair determination process.

[0064] FIG. 11 is a tenth explanatory diagram illustrating an example of experimental results in an embodiment. FIG. 11 shows that the firing propagation times from the firing electrodes D1, D2, and D3 shown in FIG. 6 to the target electrode were 4.6 ms, 2.9 ms, and 2.2 ms, respectively. The firing electrodes D1, D2, and D3 all contact the axons of presynaptic cells. It is known that a delay of 0.5 ms to several ms occurs at the location of synaptic connection, and FIG. 11 shows that the experiment yielded reasonable results consistent with this fact.

[0065] 12 is an eleventh explanatory diagram illustrating an example of an experimental result in the embodiment. Fig. 12 shows, as an example of the P value, the P value obtained for each of the firing electrodes indicated by symbols D1, D2, and D3 in Fig. 11. The P value was obtained according to the following formula (1).

[0066]

[0067] X represents a random variable. In the experiment, the physical quantity represented by the random variable X was specifically the number of firings. In the experiment, the parameter λ was specifically the product of the number of samples of the measurement data and the occurrence probability.

[0068] FIG. 13 is a twelfth explanatory diagram illustrating an example of experimental results in an embodiment. In FIG. 13, the horizontal axis indicates the number of occurrences of spike signal propagation from one electrode to another that took an extremely long time, and the vertical axis indicates the P value. The graph in FIG. 13 shows a Poisson distribution. In FIG. 13, for data with a value of 3 on the horizontal axis, the P value, which is the sum of the probability of occurrence three or more times, was 0.004.

[0069] In experiments, the P value indicates the possibility that the spike signals used in the experiment were generated accidentally due to noise or the like, and not due to the propagation of chemicals in neurons. The lower the P value, the more likely it is that they were not generated accidentally. The higher the P value, the less reliable the analysis results are, since they are due to spike signals that were generated accidentally. Therefore, it can be said that Figures 12 and 13 show that highly reliable estimation results were obtained in the experiment, as the P values ​​are small.

[0070] In the experiment, the reliability of the experimental results was evaluated using the P value, but the reliability of the experimental results is not limited to the P value and may be evaluated using any specified index that can evaluate the evaluation target and whose value is obtained by statistical testing based on observation result information.

[0071] In addition, the statistical test (hereinafter referred to as the "neuron test process") that obtains the value of a predetermined index that can evaluate the evaluation target based on the above observation result information may be performed by a person, or by another device other than the estimation device 2 or the control unit 21.

[0072] 14 is a diagram illustrating an example of the hardware configuration of the estimation device 2 according to an embodiment. The estimation device 2 includes a control unit 21 and executes a program. By executing the program, the estimation device 2 functions as a device including the control unit 21, an interface unit 22, and a storage unit 23.

[0073] More specifically, the processor 91 reads out a program stored in the storage unit 23 and stores the read program in the memory 92. The processor 91 executes the program stored in the memory 92, causing the estimation device 2 to function as a device including the control unit 21, the interface unit 22, and the storage unit 23.

[0074] The control unit 21 controls the operation of each functional unit included in the estimation device 2. The control unit 21 executes, for example, an estimation process. The control unit 21 acquires, for example, information stored in the memory unit 23. Specifically, the process of acquiring the information stored in the memory unit 23 is reading.

[0075] The interface unit 22 includes a communication interface for connecting the estimation device 2 to an external device. The interface unit 22 communicates with the external device via wired or wireless communication. The external device is, for example, the signal detector 1. The interface unit 22 acquires the detection result of the signal detector 1 by communicating with the signal detector 1.

[0076] The external device may be, for example, a predetermined output destination that outputs information obtained by the control unit 21. In such a case, the interface unit 22 outputs the information obtained by the control unit 21 to the external device that is the output destination. Such an external device that is the output destination outputs the received information, for example, as an image or sound, under the control of the control unit 21. If the external device that is the output destination is a storage device that can read and write information, the external device may readably record the received information. For example, a first type propagation intensity may be output to the external device that is the output destination. For example, a second type propagation intensity may be output to the external device that is the output destination. For example, the result of the neuron testing process may be obtained in the external device that is the output destination.

[0077] The interface unit 22 may be configured to include input devices such as a mouse, a keyboard, a touch panel, etc. The interface unit 22 may be configured as an interface that connects these input devices to the estimation device 2. In this way, the input devices of the interface unit 22 accept input of various information to the estimation device 2 via wired or wireless connections. Note that signals or information do not necessarily need to be input to the communication interface of the interface unit 22, but may also be input to the input devices of the interface unit 22.

[0078] The interface unit 22 outputs, for example, various types of information. The interface unit 22 includes, for example, a display device such as a CRT (Cathode Ray Tube) display, a liquid crystal display, or an organic EL (Electro-Luminescence) display, and a speaker. The interface unit 22 may be configured as an interface that connects these display devices or speakers to the estimation device 2. Therefore, the interface unit 22 may output, for example, information input to an input device of the interface unit 22 as an image or sound.

[0079] The storage unit 23 is configured using a computer-readable storage medium device (non-transitory computer-readable recording medium) such as a magnetic hard disk device or a semiconductor storage device. The storage unit 23 stores various information related to the estimation device 2. The storage unit 23 stores various information generated by the operation of the control unit 21, for example. The storage unit 23 may exist on a cloud, for example.

[0080] 15 is a flowchart showing an example of the flow of processing executed by the estimation device 2 in the embodiment. The control unit 21 executes the estimation processing (step S101).

[0081] The estimation device 2 configured in this manner executes the estimation process, which reduces the burden required to estimate the positions of synaptic connections, as described in <Effects of the estimation process>.

[0082] The estimation system 100 configured as above also includes the estimation device 2. This reduces the burden required to estimate the positions of synaptic connections.

[0083] (Modification) Note that multiple presynaptic cells may be estimated for one postsynaptic cell. This process may be performed by a person, or by a device other than the estimation device 2 or the control unit 21.

[0084] In addition, when multiple positions are estimated as positions of synaptic connections, a relationship identification process may be executed. The relationship identification process may obtain first-type propagation strengths for the multiple synaptic connections and estimate a predetermined relationship between multiple presynaptic cells and one postsynaptic cell in accordance with a predetermined rule.

[0085] The estimation device 2 may be implemented using a plurality of information processing devices communicably connected via a network. In this case, the respective functional units of the estimation device 2 may be distributed and implemented among the plurality of information processing devices.

[0086] Note that all or part of the functions of the estimation device 2 may be realized using hardware such as an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array). The program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include portable media such as flexible disks, magneto-optical disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into computer systems. The program may be transmitted via a telecommunications line.

[0087] Although an embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention.

[0088] The present invention includes the following aspects. <1> An estimation method performed by an estimation device, comprising: a control unit that executes an estimation process that is a process of estimating the positions of synaptic connections of neurons to be analyzed that contact a plurality of electrodes, or the pairs of cells that are synaptically connected, in which the estimation process estimates the synaptic connections or the positions of the pairs of cells that are synaptically connected based on electrode position information that indicates the positions of the electrodes; neurite position information that indicates the positions of neurites extending from the neurons; and clue information that indicates estimation signal pairs that satisfy at least an extraction condition that, among estimation signal pairs that are pairs of firing electrodes that are electrodes that are electrodes from which a propagated electrical signal exhibits a spike-like waveform, the difference between the timing at which firing occurs in one firing electrode of the pair and the timing at which firing occurs in the other firing electrode of the pair for firing, which is the generation of the spike-like waveform electrical signal, is longer than the time it takes for an electrical signal to travel a straight line distance between the one firing electrode and the other firing electrode at a standard axonal propagation velocity that is the velocity at which the electrical signal propagates through an axon. and a control step of executing an estimation process that is an estimation process. <2> The estimation method according to <1>, wherein a trigger electrode, which is one of the two firing electrodes of the estimation signal pair and which fires earlier than the other electrode, is located immediately below or in the vicinity of a neurite. <3> The estimation method according to <1> or <2>, wherein the neurite is a nerve axon. <4> The estimation method according to any one of <1> to <3>, wherein the control unit further performs a statistical test to obtain a value of a predetermined index that can evaluate the evaluation target.<5> In the control step, the second kind of relative inferring signal pair is a first kind of relative inferring signal pair that is a matching inferring signal pair that satisfies the extraction condition and a reaction electrode that is an electrode in the matching inferring signal pair that is an electrode in which firing occurs later than the other of the two firing electrodes of the inferring signal pair, the difference between the first kind of relative inferring signal pair satisfies an origin setting condition that is a predetermined condition, and the first kind of relative inferring signal pair is a first kind of relative inferring signal pair that is an electrode in which firing occurs earlier than the other of the two firing electrodes of the inferring signal pair, and the first kind of relative inferring signal pair is a first kind of relative inferring signal pair that is an electrode in which firing occurs earlier than the other of the two firing electrodes of the inferring signal pair, the difference between ... the difference between the first kind of relative inferring signal pair is a first kind of relative inferring signal pair that is an electrode in which firing occurs earlier than the other of the two firing electrodes of the inferring signal pair, and the first kind of relative inferring signal pair is a second kind of relative inferring signal pair that is a second kind of relative inferring signal pair that is a first kind of relative inferring signal pair that is an electrode in which firing occurs earlier than the other of the two firing electrodes of the inferring signal pair, the difference between the first kind of relative inferring signal pair is a first kind of relative inferring signal pair that is an electrode in which firing occurs earlier than the other of the the method for estimating a first type of propagation strength is performed based on presynaptic side inference signal pair information indicating presynaptic side inference signal pairs that are second-type relative inference signal pairs whose reaction electrodes contact presynaptic cells, postsynaptic side inference signal pair information indicating postsynaptic side inference signal pairs that are second-type relative inference signal pairs whose reaction electrodes contact postsynaptic cells, and firing count information indicating the firing counts per unit time of each second-type relative inference signal pair.<7> In the control step, the second kind of relative inferring signal pair is an inferring signal pair that is a matching inferring signal pair that satisfies the extraction condition and a reaction electrode that is an electrode in the matching inferring signal pair that is an electrode in the matching inferring signal pair that is ignited later than the other of the two ignition electrodes of the inferring signal pair, and the difference satisfies an origin setting condition that is a predetermined condition, and the first kind of relative inferring signal pair is an electrode in the first kind of relative inferring signal pair that is ignited earlier than the other of the two ignition electrodes of the inferring signal pair, and the first kind of relative inferring signal pair is an inferring signal pair that is ignited earlier than the other of the two ignition electrodes of the inferring signal pair, and the difference satisfies an origin setting condition that is a predetermined condition, and the second kind of relative inferring signal pair is an inferring signal pair that is ignited earlier than the other of the two ignition electrodes of the inferring signal pair, and the difference satisfies an origin setting condition that is a predetermined condition, The estimation method according to any one of <1> to <4>, wherein a second-type propagation strength acquisition process is performed to acquire, as a second-type propagation strength, a ratio expressed as an average value of the number of firings per unit time of the postsynaptic side estimation signal pair to an average value of the number of firings per unit time of the presynaptic side estimation signal pair, based on presynaptic side estimation signal pair information indicating a presynaptic side estimation signal pair that is a second-type relative estimation signal pair whose reaction electrode contacts a nerve axon of a presynaptic cell, postsynaptic side estimation signal pair information indicating a postsynaptic side estimation signal pair that is a second-type relative estimation signal pair whose reaction electrode contacts a postsynaptic cell, and firing count information indicating the number of firings per unit time of each second-type relative estimation signal pair.<9> An estimation device including: a control unit that executes an estimation process that estimates the positions of synaptic connections of neurons that are in contact with a plurality of electrodes, or the pairs of cells that are synaptically connected, in which the estimation process estimates the positions of the synaptic connections or the pairs of cells that are synaptically connected, based on: electrode position information that indicates the positions of the electrodes; neurite position information that indicates the positions of neurites extending from the neurons; and clue information that indicates estimation signal pairs that are pairs of firing electrodes that are electrodes from which a propagated electrical signal exhibits a spike-like waveform, and that satisfy at least the extraction condition that the difference between the timing at which firing occurs at one firing electrode of the pair and the timing at which firing occurs at the other firing electrode of the pair, for firing, which is the generation of the electrical signal with the spike-like waveform, is longer than the time it takes for an electrical signal to travel a straight line distance between the one and the other at a standard axonal propagation velocity, which is the velocity at which the electrical signal propagates through an axon. <10> An estimation system comprising: a plurality of electrodes; and a control unit that executes an estimation process that estimates synaptic connections of neurons to be analyzed that are in contact with the plurality of electrodes, or the positions of the synaptically connected cell pairs, in which the estimation process estimates the positions of the synaptic connections or the synaptically connected cell pairs based on electrode position information that indicates the positions of the electrodes, neurite position information that indicates the positions of neurites extending from the neurons, and clue information that indicates estimation signal pairs that are pairs of firing electrodes that are electrodes from which a propagated electrical signal exhibits a spike-like waveform, and that satisfy at least an extraction condition that the difference between the timing at which firing occurs at one firing electrode of the pair and the timing at which firing occurs at the other firing electrode of the pair for firing, which is the generation of the spike-like waveform electrical signal, is longer than the time it takes for an electrical signal to travel a straight line between the one firing electrode and the other firing electrode at a standard axonal propagation velocity, which is the velocity at which the electrical signal propagates through an axon. <11> A program for causing a computer to function as the estimation device according to <9>. <12> A program for causing a computer to function as the estimation system according to <10>.

[0089] 100 Estimation system 1 Signal detector 2 Estimation device 11 Electrode 21 Control unit 22 Interface unit 23 Storage unit 91 Processor 92 Memory

[0090] J. BARTRAM, M. SCHROTER, S. RONCHI, V. EMMENEGGER, J. MULLER, A. HIERLEMANN, “Mechanisms of homeostatic synaptic plasticity”, Neuroscience 2018 Session 037, 037.11 / E1

Claims

1. A control unit that executes an estimation process to estimate the synaptic connection positions of a neuron to be analyzed that contacts a plurality of electrodes, or the pair of cells that are synaptically connected, and the estimation process includes: electrode position information indicating the position of each of the electrodes; neurite position information indicating the position of a neurite extending from the neuron; and, among a pair of estimation signals that are pairs of firing electrodes that are electrodes that transmit a spike-shaped waveform, timing at which firing occurs at one firing electrode of the pair and timing at which firing occurs at the other firing electrode of the pair, for firing, which is the generation of an electrical signal with a spike-shaped waveform. and a control step of executing an estimation process that estimates the positions of synaptic connections of neurons to be analyzed that contact multiple electrodes, or the pairs of synaptically connected cells, based on clue information that indicates a pair of estimation signals that at least satisfy an extraction condition that the difference in timing between the pair of estimation signals is longer than the time it takes for an electrical signal to travel a straight-line distance between the one and the other at a standard axonal propagation speed, which is the speed at which an electrical signal propagates through an axon.

2. The estimation method according to claim 1, wherein the trigger electrode, which is one of the two firing electrodes of the estimation signal pair and which fires earlier than the other electrode, is located immediately below or in the vicinity of the neurite.

3. The estimation method according to claim 1, wherein the neurite is a nerve axon.

4. The estimation method according to claim 1, wherein the control unit further performs a statistical test to obtain a value of a predetermined index capable of evaluating the evaluation target.

5. In the control step, a second kind of relative inferring signal pair is further selected, which is a first kind of relative inferring signal pair that is a matching inferring signal pair that satisfies the extraction condition and a reaction electrode that is an electrode in the matching inferring signal pair that is one of the two firing electrodes of the inferring signal pair that ignites later than the other electrode, and which has an origin electrode that is an electrode in the first kind of relative inferring signal pair that ignites earlier than the other electrode, the difference between which satisfies an origin setting condition that is a predetermined condition, and which has an ignition electrode of the first kind of relative inferring signal pair excluding the origin electrode as a trigger electrode.

2. The estimation method according to claim 1, further comprising: a first-type propagation strength acquisition process for acquiring, as a first-type propagation strength, a ratio expressed as an average value of the number of firings per unit time of the postsynaptic side estimation signal pair to an average value of the number of firings per unit time of the presynaptic side estimation signal pair, based on: presynaptic side inference signal pair information indicating presynaptic side inference signal pairs that are second-type relative inference signal pairs whose reaction electrodes contact presynaptic cells; postsynaptic side inference signal pair information indicating postsynaptic side inference signal pairs that are second-type relative inference signal pairs whose reaction electrodes contact postsynaptic cells, among the second-type relative inference signal pairs; and firing count information indicating the number of firings per unit time of each second-type relative inference signal pair.

6. The estimation method according to claim 5, wherein the origin setting condition is that the firing propagation time is longer than that of other first-kind relative inferring signal pairs.

7. In the control step, the second kind of relative inferring signal pair is a first kind of relative inferring signal pair that is a matching inferring signal pair that satisfies the extraction condition and a reaction electrode that is an electrode in the matching inferring signal pair that ignites later than the other of the two ignition electrodes of the inferring signal pair, and the second kind of relative inferring signal pair is a first kind of relative inferring signal pair that satisfies the origin setting condition, and the difference satisfies the predetermined condition, that is, an origin electrode that is an electrode in the first kind of relative inferring signal pair that ignites earlier than the other of the two ignition electrodes of the inferring signal pair, and the second kind of relative inferring signal pair is a first kind of relative inferring signal pair that ignites earlier than the other of the two ignition electrodes of the inferring signal pair, and the first ... second kind of relative inferring signal pair is a second kind of relative inferring signal pair that is a first kind of relative inferring signal pair that ignites earlier than the other of the two ignition electrodes of the inferring signal pair, and the difference satisfies the predetermined condition, that is, an origin setting condition, 2. The estimation method according to claim 1, wherein a second-type propagation strength acquisition process is performed to acquire, as a second-type propagation strength, a ratio expressed as an average value of the number of firings per unit time of the postsynaptic side estimation signal pair to an average value of the number of firings per unit time of the presynaptic side estimation signal pair, based on: presynaptic side estimation signal pair information indicating presynaptic side estimation signal pairs that are second-type relative estimation signal pairs whose reaction electrodes contact a nerve axon of a presynaptic cell; postsynaptic side estimation signal pair information indicating postsynaptic side estimation signal pairs that are second-type relative estimation signal pairs whose reaction electrodes contact a postsynaptic cell, among the second-type relative estimation signal pairs; and firing count information indicating the number of firings per unit time of each second-type relative estimation signal pair.

8. The estimation method according to claim 7, wherein the origin setting condition is that the firing propagation time is longer than that of other first-kind relative inferring signal pairs.

9. An estimation device comprising: a control unit that executes an estimation process that estimates the positions of synaptic connections of neurons to be analyzed that are in contact with a plurality of electrodes, or the pairs of cells that are synaptically connected, in which the estimation process estimates the positions of the synaptic connections or the pairs of cells that are synaptically connected based on: electrode position information that indicates the positions of each of the electrodes; neurite position information that indicates the positions of neurites extending from the neurons; and clue information that indicates estimation signal pairs that are pairs of firing electrodes that are electrodes from which a propagated electrical signal exhibits a spike-like waveform, and that satisfy at least the extraction condition that the difference between the timing at which firing occurs at one firing electrode of the pair and the timing at which firing occurs at the other firing electrode of the pair for firing, which is the generation of the spike-like waveform electrical signal, is longer than the time it takes for an electrical signal to travel a straight line distance between one firing electrode and the other firing electrode at the standard axon propagation velocity, which is the velocity at which the electrical signal propagates through the axon.

10. An estimation system comprising: a plurality of electrodes; and a control unit that executes an estimation process that estimates the synaptic connections of neurons to be analyzed that are in contact with the plurality of electrodes, or the positions of pairs of synaptically connected cells, in which the estimation process estimates the positions of the synaptic connections or the pairs of synaptically connected cells based on electrode position information that indicates the positions of the electrodes, neurite position information that indicates the positions of neurites extending from the neurons, and clue information that indicates estimation signal pairs that are pairs of firing electrodes that are electrodes from which a propagated electrical signal exhibits a spike-like waveform, and that satisfy at least the extraction condition that the difference between the timing at which firing occurs at one firing electrode of the pair and the timing at which firing occurs at the other firing electrode of the pair for firing, which is the generation of the spike-like waveform electrical signal, is longer than the time it takes for an electrical signal to travel a straight line between one firing electrode and the other firing electrode at a standard axonal propagation velocity, which is the speed at which the electrical signal propagates through an axon.

11. A program for causing a computer to function as the estimation device according to claim 9.

12. A program for causing a computer to function as the estimation system according to claim 10.

Images