CN104881542B - The component method for arranging and system of biochemical reaction detection device on micro chip - Google Patents
The component method for arranging and system of biochemical reaction detection device on micro chip Download PDFInfo
- Publication number
- CN104881542B CN104881542B CN201510279070.XA CN201510279070A CN104881542B CN 104881542 B CN104881542 B CN 104881542B CN 201510279070 A CN201510279070 A CN 201510279070A CN 104881542 B CN104881542 B CN 104881542B
- Authority
- CN
- China
- Prior art keywords
- probability distribution
- biochemical reaction
- physical position
- component
- layout
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Landscapes
- Apparatus Associated With Microorganisms And Enzymes (AREA)
- Other Investigation Or Analysis Of Materials By Electrical Means (AREA)
Abstract
The present invention provides the component method for arranging and system of biochemical reaction detection device on a kind of micro chip, the described method includes：It generates N number of component physical location layout and generates the component physical location and be laid out corresponding probability distribution parameter；Calculate the biochemical reaction deadline of each component physical location layout；It chooses and the probability distribution parameter of k shortest biochemical reaction deadlines corresponding component physical location layout obtains updated probability distribution parameter before updating；New sample is generated according to the updated probability distribution parameter；It repeats m times until the probability distribution parameter of newly generated component physical location layout provides definite component physical location layout.The method of the invention can realize the time for reducing entire biochemical reaction detection process by the physical location and its connection path for rationally designing component.
Description
Technical Field
The invention belongs to the field of biochemical reaction detection, and particularly relates to a component arrangement method and a component arrangement system of a biochemical reaction detection device on a microchip.
Background
In the conventional biochip, because on-chip reaction resources are limited, whether the component layout is reasonable directly affects the detection completion time, and in order to better utilize the limited on-chip resources for different detections so as to reduce the detection time, a reasonable physical position layout of the components and a layout scheme of connection paths thereof need to be set.
Disclosure of Invention
The invention provides a component arrangement method and a component arrangement system of a biochemical reaction detection device on a microchip, which are used for reducing the time of the whole biochemical reaction detection process by reasonably designing the physical position of a component and a connection path thereof.
The first aspect of the present invention provides a method for arranging components of a biochemical reaction detecting apparatus on a microchip, comprising:
generating N component physical position layouts and generating probability distribution parameters corresponding to the component physical position layouts;
calculating the biochemical reaction completion time of the physical position layout of each assembly;
selecting and updating probability distribution parameters of the component physical position layout corresponding to the first k shortest biochemical reaction completion times to obtain updated probability distribution parameters;
generating a new sample according to the updated probability distribution parameter;
repeatedly executing the generating layout, the calculating time, the selecting and updating the probability distribution parameters of the component physical position layout for m times until the probability distribution parameters of the newly generated component physical position layout give the determined component physical position layout;
wherein N is more than 1,1 and k are woven together and N is more than or equal to 1.
In a second aspect of the present invention, there is provided a component arrangement system for a biochemical reaction detecting apparatus on a microchip, comprising:
a position layout module for generating N component physical position layouts and generating probability distribution parameters corresponding to the component physical position layouts, calling the results of biochemical completion time calculation performed by the routing layout module and accordingly selecting and updating the probability distribution parameters of the component physical position layouts corresponding to the former k shortest biochemical reaction completion times to obtain updated probability distribution parameters, generating new samples according to the updated probability distribution parameters, and repeatedly executing the generating layout, the biochemical reaction completion time calculation results of the calling routing layout module, the selecting and updating the probability distribution parameters of the component physical position layouts until the probability distribution parameters of the newly generated component physical position layouts give a determined component physical position layout; wherein N is more than 1,1 and k are woven together, and m is more than or equal to 1;
and the routing layout module is used for calculating the biochemical reaction completion time of the physical position layout of each component.
The beneficial effects of the invention are as follows:
the component arrangement method of the biochemical reaction detection device on the microchip provided by the invention has the advantages that the physical position layout variable of the biochemical reaction component to be optimized is represented by probability distribution, the probability distribution of the component layout variable is optimized by minimizing the reaction completion time, the layout sample set is updated according to the reaction time under the physical position layout of the component, the optimization process is iterated repeatedly until the probability distribution corresponding to the layout sample set gives a determined layout, the reasonable design of the physical position and the connection path of the component is realized, and the time of the whole biochemical reaction detection process can be reduced.
Drawings
FIG. 1 is a flow chart of a first embodiment of a method for arranging components of a biochemical reaction detecting apparatus on a microchip according to the present invention;
FIG. 2 is a flowchart of a PCR test sample in a first embodiment of a method for arranging elements of the on-chip biochemical reaction detection apparatus according to the present invention;
FIG. 3 is a schematic diagram showing the structure of a PCR biochemical experiment in a first embodiment of the method for arranging the elements of the on-chip biochemical reaction detection apparatus according to the present invention;
FIG. 4 is a schematic view of a maze algorithm in a first embodiment of a method for arranging elements of the on-chip biochemical reaction detecting device according to the present invention;
FIG. 5 is a diagram showing an initial layout of PCR biochemical experiments in a first embodiment of a method for arranging elements of the on-chip biochemical reaction detection apparatus according to the present invention;
FIG. 6 is a diagram showing the initial exchange probability distribution of a PCR biochemical experiment in a first embodiment of the method for arranging components of the on-chip biochemical reaction detecting apparatus according to the present invention;
FIG. 7 is a diagram showing an initial exchange probability distribution of PCR biochemical experiments after one iteration in the first embodiment of the method for arranging elements of the on-chip biochemical reaction detection apparatus according to the present invention;
FIG. 8 is a block diagram of a component arrangement system of a biochemical reaction detecting apparatus on a microchip according to a first embodiment of the present invention.
Detailed Description
FIG. 1 is a flowchart of a first embodiment of a method for arranging components of a biochemical reaction detecting device on a microchip according to the present invention, and FIG. 2 is a flowchart of a first embodiment of a method for arranging components of a biochemical reaction detecting device on a microchip according to the present invention, as shown in FIGS. 1 and 2, the method for arranging components of a biochemical reaction detecting device on a microchip according to the present invention comprises:
s101, generating N component physical position layouts and generating probability distribution parameters corresponding to the component physical position layouts; the biochemical reaction components include but are not limited to input or output liquid storage tanks, dilution reaction tanks (channels); the routing layout includes, but is not limited to, a routing method for conducting biochemical reaction reagents or biochemical reactions in channels disposed between reservoirs; the assembly placement method may be applied on including but not limited to biochips;
preferably, the generating N component physical location layouts (samples) and generating the probability distribution parameters corresponding to the component physical location layouts includes:
randomly generating N first-time component physical position layouts and generating first-time probability density parameters corresponding to the component physical position layouts;
alternatively, the first and second electrodes may be,
generating N (i + 1) th sub-assembly physical position layouts according to the updated probability distribution parameters of the ith sub-assembly physical position layout, preferably comprising: and generating N (i + 1) th sub-assembly physical position layouts by Monte Carlo simulation according to the updated probability distribution parameters of the ith sub-assembly physical position layout.
S102, calculating the biochemical reaction completion time of the physical position layout of each assembly; during specific calculation, the functional relationship between the biochemical reaction completion time of the physical position layout of each component and the physical position layout of the component can be expressed by the biochemical reaction completion time of the physical position layout of each component and a probability distribution parameter of the physical position layout of the component; for simplifying the description, the probability distribution parameters of the component physical location layout of the biochemical reaction completion time function are simply referred to as the probability distribution parameters of the component physical location layout hereinafter; preferably, the calculating the biochemical reaction completion time of the physical position layout of each component includes:
calculating the biochemical reaction completion time in a plurality of routing layouts corresponding to each component physical position layout according to a maze algorithm;
selecting the minimum biochemical reaction completion time as the biochemical reaction completion time of the physical position layout of the components, and simultaneously selecting the route layout corresponding to the minimum biochemical reaction completion time as the connection scheme of each component in the physical position layout of the components;
preferably, the following example of the layout of a specific module is used to illustrate how the maze algorithm is used to calculate the completion time of the biochemical reaction:
the design of the biochip consists of two parts, namely architecture synthesis and physical synthesis. Architecture synthesis enables the allocation of limited on-chip resources to various basic operations, each of which has a one-to-one correspondence to each component (e.g., mixer and heater, etc.) in the actual chip as shown in fig. 2. The whole biochip is composed of a series of components and micro valves. The assembly completes various operations required by biochemical tests, and the micro-valve controls the transmission of reagents in the biochip according to a set path. Physical synthesis involves the placement of the various components on the chip and the channel routing that connects the various components.
Architecture synthesis (architecture synthesis) is used to determine a one-to-one mapping of each operation to a specific chip component. This synthesis can be implemented by a simple greedy algorithm. The greedy algorithm assigns finite components one by one according to the topological ordering of the biochemical test flow chart. A simple architectural complex example is shown in fig. 7. In topological order, operations 1,2,4,5,7,8,10,11, without a preceding operation, may be given to the on-chip components. Due to the resource limitations of the chip (only 6 input reservoirs), operations 1,2,4,5,7,8 are assigned to inputs 1,2,3,4,5,6, respectively, and completed in time period 1. Operations 10,11 are assigned to inputs 5 and 6, and are completed at time period 2. Operations 3,6 and 9 are completed by being assigned to diluter 1,2 and 3, respectively, during which time period all three diluters on the chip are used. During time period 3, operations 17,18, and 19 are assigned to outputs 1,2, and 3, respectively. At this point diluter resources are released and operation 12 is allowed to proceed at dilution 3. Resource allocation thereafter and so on. When the on-chip resources are insufficient to satisfy all operations that are currently possible, a portion of the currently executable operations will be delayed until the next time period for execution.
As shown in fig. 4, after the layout of the biochip is determined, devices in a sequential relationship need to be connected by channels. The process of determining the inter-device wiring is very similar to the VLSI wiring process, with the only difference being: the wires in the VLSI circuits cannot be crossed, while the wires in the biochips allow crossing. When crossover occurs, conduction of the reagents cannot occur simultaneously at the crossover. This means an extension of the test completion time. Therefore, channel crossings should be avoided.
The maze routing algorithm in VLSI routing can be used to solve the routing problem in biochips. For each pair of interconnected devices, the maze routing algorithm may find the shortest path with the smallest number of routing intersections. The algorithm propagates the tag routing cost outward from the starting point until every point around the target is tagged. At some iteration, every neighbor node of the node visited by the last iteration is marked. An example of a simple maze routing algorithm (maze routing algorithm) is shown in fig. 4.
In this case, the input 5 needs to be connected to the dilution 3. The existing connections include input 1 and dilution 1, and dilution 1 and dilution 3. The labels in each iteration are represented by the same color. In the first iteration, the positions above and below input 5 are marked 1, indicating that the cost of routing to this from input 5 is 1. In the second iteration, the propagation is outward starting from positions above and below the input 5, 3 positions being marked 2 in the figure. To avoid channel crossing, the routing node costs 2 at the locations already occupied by other channels. E.g., the position below input 1, the cost of propagating to this from input 5 is 2+2=4. When the marking is completed, the optimal path can be found by backtracking from the target point. The lowest cost to reach dilution 3 is 7. From this point, the neighboring lowest cost node is sought each time. In the example of fig. 4, the lowest cost neighboring node of the node having the cost of 7 is 6. And so on until input 5 is reached. The final path is marked by a red line.
A brief description of biochip timing analysis follows:
the timing analysis of biochips is very similar to architectural synthesis. Different from architecture synthesis, timing analysis needs to consider the time of channel conduction reagent and the delay caused by channel time sharing. For example, in FIG. 4, the channels of inputs 1-Dilute 1 cross the channels of inputs 5-Dilute 3. Therefore, the two channels cannot operate in the same time period. This results in input 5 and input 6 needing to be moved from the first time period to the second time period. The operations after input 5 and input 6 are therefore straightforward.
The timing analysis may be implemented by a simple greedy algorithm. Similar to architectural synthesis, the operations are processed sequentially in the topological order of the flow diagram. The completion time of each operation is defined by the following equation:
T prev is the latest completion time, T, in the preamble operation curr Is the operating time of the current operation, T d Due to the delay caused by channel time sharing. The last completed operation defines the completion time of the entire biochemical test.
S103, selecting and updating probability distribution parameters of the component physical position layout corresponding to the former k shortest biochemical reaction completion times to obtain updated probability distribution parameters;
preferably, the selecting and updating the probability distribution parameters of the component physical location layout corresponding to the first k shortest biochemical reaction completion times to obtain updated probability distribution parameters includes:
selecting probability distribution parameters of the component physical position layout corresponding to the first k shortest biochemical reaction completion times;
updating the k selected probability distribution parameters according to the probability distribution parameters obtained by minimizing the biochemical reaction completion time by a cross entropy algorithm; preferably, the method comprises the following steps:
s1031, condition (1) of obtaining the first intermediate function l → 0 according to the minimum biochemical reaction completion time:
s1032, obtaining a second intermediate function according to the condition of the first intermediate function l → 0And approximation function g x (x) Condition (2) that the cross entropy (Kullback-Leibler Diffenrence, KL divergence) of (c) is minimum:
s1033, determining a probability distribution parameter v' under the condition of minimum cross entropy:
s1034, updating the probability distribution parameter v by using v';
wherein f is x (x; v) represents the time of completion of the biochemical reaction versus the physical location of the assembly x (or x) i ) Is simply denoted as f (X), v represents the probability that the component physical location layout X corresponds toDistribution parameter, g x (x i ) Denotes f x (x; v), gamma represents the minimum value of the completion time of the biochemical reaction, P represents the probability, I represents the probability of the biochemical reaction f(X)≤γ (= 1) indicating function, I f(X)≤γ ＝1、I| f(X)>γ =0,n denotes the resulting physical location layout x of the component i The number of (2);
in order to more clearly illustrate the implementation of the method, we give detailed derivation processes of S1031 to S1034 as follows:
for the objective function f (X) defined on X ∈ X, all X ∈ X are independent. Defining the distribution of X ∈ X following a probability density function f x (x; v) wherein the vector v of finite dimensions is f x (x; v). The goal of the cross-entropy method is to determine whether there is a constant γ such that f (X). Ltoreq.γ becomes a small probability event, i.e.:
l＝P(f(X)≤γ)＝E[I {f(X)≤γ} ]＝∫I {f(X)≤γ} f x (x；v) (2)；
since we need to analyze the case where l becomes small, we introduce another probability density function g (x) that, for all x,applying the definition of g (x), l can be expressed as:
due to X 1 ,…X N All are random vectors where g (x) is independent of each other, and the significance sampling of l can be estimated as:
we need to find a specific g (x) to minimize l'. The density of the random variable X at f (X). Ltoreq.gamma is expressed as:
we need to minimize the Kullback-Leibler difference between g and g', i.e. to minimize the following equation:
we need to find a specific v such that ^ g '(x) lng (x) dx is minimized, i.e. minimized ^ g' (x) lnf (x: v) dx, i.e. the following formula is maximized
That is to say
max v D(v)＝max v E u I {f(X)≤γ} lnf(x；v) (9)；
The importance samples are used to obtain the importance samples,
wherein w is an arbitrary parameter, isA probability ratio. Thus, v' can be estimated by
Now the random variable X ∈ X obeys the probability density function f (X; w).
To solve the biochip layout, we propose an algorithm based on the cross entropy method. The algorithm iteratively optimizes from an initial layout of the biochip until the layout cannot be further optimized. For simplicity, the originalThe starting biochip is logically divided into N (N)>＝N 0 ) A grid. Wherein N is 0 It is the total number of devices on the biochip, which can be obtained by architecture synthesis. An initial PCR layout is shown in figure 5. As can be seen, each device occupies a single net.
The samples in this algorithm specify the swapping of the positions of two different devices. For example, input 1 is swapped in position with input 2. The effect of device switching is characterized by the switching score. The exchange score is used to evaluate the effect of the current exchange and is linearly inversely proportional to the trial completion time resulting from the current layout. For each exchange, wiring synthesis and time sequence simulation are completed according to the current biochip layout, and the completion time of the whole biochemical test is obtained. Different placement plans may result in different routing plans that affect completion time. A good device swap is beneficial to reduce the test completion time.
The switching choices for each pair of devices obey a particular probability distribution. We sample the normal distribution in the algorithm implementation. In each iteration, we randomly choose n samples (device swap). For each sample, a random number is generated according to its distribution, and if the number is greater than a constant threshold, the sample is selected. For each selected sample, we apply wiring planning and timing analysis to its newly generated layout. After the exchange score is obtained, the probability distribution of the exchange of the device is updated according to the exchange score. We select the best k out of all the selected samples and update their probability distributions, i.e. increase the mean and decrease the variance of the normal distribution. This will result in these device swaps having more opportunity to be selected in the next iteration. Before this iteration is finished, the best sample is executed. The newly generated layout becomes the base layout for the next iteration, on which all device swaps are generated.
Before the first iteration starts, the device swap probability distribution of the PCR test sample is shown in FIG. 6. I.e. each pair of exchanges has the same chance to be selected. After one iteration, the probability distribution of each exchange combination is shown in fig. 7. Let k =4, the best k samples are updated with probability distributions so that they have a greater probability of being selected in the next iteration.
The iterative process of the algorithm continues until the layout cannot be further optimized. Each selected combination of exchanges does not result in faster trial completion times in the timing simulation. This indicates that the algorithm has been unable to continue optimizing the existing layout. We consider the existing layout to be already close to optimal.
The following illustrates a specific manner of obtaining v 'update according to the update result v' (3) of the probability distribution parameters given in S1031 to S1034 after a specific probability density parameter function is given, as follows:
preferably, the updating the k selected probability distribution parameters according to the probability distribution parameters obtained by minimizing the biochemical reaction completion time by the cross entropy algorithm includes:
setting the physical position parameter in the probability distribution parameter of the component physical position layout corresponding to the first k shortest biochemical reaction completion times as the average value of the physical position parameter in the probability distribution parameter of the component physical position layout corresponding to the first k shortest biochemical reaction completion times, and reducing the value of the physical position variance parameter in the probability distribution parameter of the component physical position layout corresponding to the first k shortest biochemical reaction completion times;
s104, generating a new sample according to the updated probability distribution parameters;
s105, repeatedly executing the generation layout, the calculation time and the probability distribution parameters of the selected and updated component physical position layout for m times until the probability distribution parameters of the newly generated component physical position layout give a determined component physical position layout; wherein N is more than 1,1 and k are woven together, and m is more than or equal to 1; here, the result of such iterative multiple optimizations is.
The component arrangement method of the biochemical reaction detection device on the microchip provided by the invention has the advantages that the physical position layout variable of the biochemical reaction component to be optimized is represented by probability distribution, the probability distribution of the component layout variable is optimized by minimizing the reaction completion time, the layout sample set is updated according to the reaction time under the physical position layout of the component, the optimization process is iterated repeatedly until the probability distribution corresponding to the layout sample set gives a determined layout, the reasonable design of the physical position and the connection path of the component is realized, and the time of the whole biochemical reaction detection process can be reduced.
FIG. 1 is a flowchart of a first embodiment of a component arrangement system of a biochemical reaction detecting apparatus on a microchip according to the present invention, and FIG. 2 is a flowchart of a first embodiment of a component arrangement system of a biochemical reaction detecting apparatus on a microchip according to the present invention, as shown in FIGS. 1 and 2, the component arrangement system of a biochemical reaction detecting apparatus on a microchip according to the present invention comprises:
a position layout module 21, configured to generate N component physical position layouts and generate probability distribution parameters corresponding to the component physical position layouts, invoke a result of biochemical completion time calculation performed by the routing layout module, select and update probability distribution parameters of the component physical position layout corresponding to the first k shortest biochemical reaction completion times according to the result to obtain updated probability distribution parameters, generate a new sample according to the updated probability distribution parameters, and repeatedly execute the generation layout, the biochemical reaction completion time calculation result of the invocation routing layout module, and the probability distribution parameters of the component physical position layouts until the probability distribution parameters of the newly generated component physical position layouts give a determined component physical position layout; wherein N is more than 1,1 and k are woven together, and m is more than or equal to 1;
and the routing layout module 22 is used for calculating the biochemical reaction completion time of the physical position layout of each component.
The biochemical reaction components include but are not limited to input or output liquid storage tanks, dilution reaction tanks (channels); the routing layout includes, but is not limited to, a routing method for conducting biochemical reaction reagents or biochemical reactions occurring in channels disposed between reservoirs; the assembly placement method may be applied on including but not limited to biochips;
preferably, the position layout module 21 is specifically configured to:
randomly generating N first sub-component physical position layouts, generating N (i + 1) th sub-component physical position layouts corresponding to the component physical position layouts, selecting the probability distribution parameters of the component physical position layouts corresponding to the first k shortest biochemical reaction completion times, updating the selected k probability distribution parameters according to the probability distribution parameters obtained by minimizing the biochemical reaction completion time condition through a cross entropy algorithm, and generating N (i + 1) th sub-component physical position layouts through Monte Carlo simulation according to the updated probability distribution parameters of the ith sub-component physical position layouts.
Preferably, the position layout module 21 is specifically configured to:
updating the selected k probability distribution parameters according to probability distribution parameters obtained by minimizing the biochemical reaction completion time by a cross entropy algorithm comprises:
the condition (4) that the first intermediate function l → 0 is obtained from the biochemical reaction completion time being minimum:
obtaining a second intermediate function according to the condition of the first intermediate function l → 0And approximation function g x (x) Condition (7) that the cross entropy (Kullback-Leibler diffrence, KL divergence) of (c) is minimum:
determining a probability distribution parameter v' (11) under the condition of minimum cross entropy:
and updating the probability distribution parameter v with v';
wherein f is x (x; v) represents the time of completion of the biochemical reaction versus the physical location of the assembly x (or x) i ) Is abbreviated as f (X), v represents the probability distribution parameter corresponding to the physical position layout X of the component, g x (x i ) Denotes f x (x; v), gamma represents the minimum value of the completion time of the biochemical reaction, P represents the probability, I represents the probability of the biochemical reaction f(X)≤γ (= 1) indicating function, I f(X)≤γ ＝1、I| f(X)>γ =0,n denotes the resulting physical location layout x of the component i The number of (2).
Preferably, the routing layout module 22 is specifically configured to calculate biochemical reaction completion times in a plurality of routing layouts corresponding to each component physical location layout according to a maze algorithm;
correspondingly, the position layout module 21 is specifically configured to select a minimum biochemical reaction completion time as the biochemical reaction completion time for the physical position layout of the components, and select a routing layout corresponding to the minimum biochemical reaction completion time as a connection scheme for each component in the physical position layout of the components.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and these modifications or substitutions do not depart from the spirit of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for arranging components of a biochemical reaction detecting apparatus on a microchip, comprising:
generating N component physical position layouts and generating probability distribution parameters corresponding to the component physical position layouts;
calculating the biochemical reaction completion time of the physical position layout of each component;
selecting and updating probability distribution parameters of the component physical position layout corresponding to the first k shortest biochemical reaction completion times to obtain updated probability distribution parameters;
generating a new sample according to the updated probability distribution parameter;
repeatedly executing the generation layout, the calculation time and the probability distribution parameters of the selected and updated component physical position layout m times until the probability distribution parameters of the newly generated component physical position layout give the determined component physical position layout;
wherein N is more than 1,1 and k are woven together and N is more than or equal to 1.
2. The method of claim 1, wherein the generating N component physical location layouts and generating probability distribution parameters corresponding to the component physical location layouts comprises:
randomly generating N first sub-component physical position layouts and generating first sub-probability density parameters corresponding to the component physical position layouts;
alternatively, the first and second liquid crystal display panels may be,
generating N (i + 1) th sub-assembly physical position layouts according to the updated probability distribution parameters of the ith sub-assembly physical position layout;
correspondingly, the selecting and updating the probability distribution parameters of the component physical position layout corresponding to the first k shortest biochemical reaction completion times to obtain updated probability distribution parameters includes:
selecting probability distribution parameters of component physical position layout corresponding to the first k shortest biochemical reaction completion times;
and updating the selected k probability distribution parameters according to the probability distribution parameters obtained by minimizing the biochemical reaction completion time by a cross entropy algorithm.
3. The method of claim 2, wherein the generating N (i + 1) th sub-assembly physical location layouts according to the updated probability distribution parameters of the ith sub-assembly physical location layout comprises:
and generating N (i + 1) th sub-assembly physical position layouts by Monte Carlo simulation according to the updated probability distribution parameters of the ith sub-assembly physical position layout.
4. The method for arranging the components of a biochemical reaction detecting apparatus on a microchip according to claim 2, wherein the updating the selected k probability distribution parameters based on the probability distribution parameters obtained by the condition that the cross entropy algorithm minimizes the completion time of the biochemical reaction comprises:
the condition (4) that the first intermediate function l → 0 is obtained from the biochemical reaction completion time being minimum:
obtaining a second intermediate function according to the condition of the first intermediate function l → 0And approximation function g x (x) Condition (7) of minimum cross entropy of (a):
determining a probability distribution parameter v' (11) under the condition of minimum cross entropy:
updating the probability distribution parameter v by v';
wherein f is x (x; v) represents the time of completion of the biochemical reaction versus the physical location of the assembly x (or x) i ) Is abbreviated as f (x), v represents the probability distribution parameter corresponding to the physical position layout x of the component, g x (x i ) Denotes f x (x; v), gamma represents the minimum value of the completion time of the biochemical reaction, P represents the probability, I represents the probability of the biochemical reaction f(x)≤γ =1 denotes an indicating function, I- f(x)≤γ ＝1、I| f(x)＞γ =0,n denotes the resulting physical location layout x of the component i The number of the cells.
5. The method of claim 4, wherein the probability density parameter is a multi-dimensional Gaussian distribution parameter vector, and the updating the selected k probability distribution parameters according to the probability distribution parameters obtained by the cross entropy algorithm minimizing the biochemical reaction completion time comprises:
setting the physical position parameter in the probability distribution parameter of the component physical position layout corresponding to the first k shortest biochemical reaction completion times as the average value of the physical position parameter in the probability distribution parameter of the component physical position layout corresponding to the first k shortest biochemical reaction completion times, and reducing the value of the physical position variance parameter in the probability distribution parameter of the component physical position layout corresponding to the first k shortest biochemical reaction completion times.
6. The method of claim 1, wherein the calculating the biochemical reaction completion time for the physical location layout of each of the modules comprises:
calculating the biochemical reaction completion time in a plurality of routing layouts corresponding to the physical position layout of each component according to a maze algorithm;
and selecting the minimum biochemical reaction completion time as the biochemical reaction completion time of the physical position layout of the components, and simultaneously selecting the route layout corresponding to the minimum biochemical reaction completion time as the connection scheme of each component in the physical position layout of the components.
7. A component placement system for a biochemical reaction detecting apparatus on a microchip, comprising:
the position layout module is used for generating N component physical position layouts and generating probability distribution parameters corresponding to the component physical position layouts, calling the results of biochemical completion time calculation executed by the routing layout module, selecting and updating the probability distribution parameters of the component physical position layouts corresponding to the former k shortest biochemical reaction completion times according to the results to obtain updated probability distribution parameters, generating new samples according to the updated probability distribution parameters, and repeatedly executing the generating layout, calling the biochemical reaction completion time calculation results of the routing layout module and selecting and updating the probability distribution parameters of the component physical position layouts until the probability distribution parameters of the newly generated component physical position layouts give determined component physical position layouts; wherein N is more than 1,1 and k are woven together, and m is more than or equal to 1;
and the routing layout module is used for calculating the biochemical reaction completion time of the physical position layout of each component.
8. The system of claim 7, wherein the placement module is specifically configured to:
randomly generating N first-time component physical position layouts and generating N (i + 1) th-time component physical position layouts corresponding to the component physical position layouts, or generating N (i + 1) th-time component physical position layouts according to updated probability distribution parameters of the ith-time component physical position layout, selecting the probability distribution parameters of the component physical position layout corresponding to the first k shortest biochemical reaction completion times, updating the selected k probability distribution parameters according to the probability distribution parameters obtained by minimizing the condition of the biochemical reaction completion time through a cross entropy algorithm, and generating N (i + 1) th-time component physical position layouts through Monte Carlo simulation according to the updated probability distribution parameters of the ith-time component physical position layout.
9. The system of claim 8, wherein the placement module is specifically configured to:
updating the selected k probability distribution parameters according to probability distribution parameters obtained by minimizing the biochemical reaction completion time by a cross entropy algorithm comprises:
the condition (4) for obtaining the first intermediate function l → 0 according to the minimum biochemical reaction completion time:
obtaining a second intermediate function according to the condition of the first intermediate function l → 0And approximation function g x (x) Condition (7) of minimum cross entropy of (a):
determining a probability distribution parameter v' (11) under the condition of minimum cross entropy:
and updating the probability distribution parameter v with v';
wherein f is x (x; v) represents the time of completion of the biochemical reaction versus the physical location of the assembly x (or x) i ) Is abbreviated as f (x), v represents the probability distribution parameter corresponding to the physical position layout x of the component, g x (x i ) Denotes f x (x; v), gamma represents the minimum value of the completion time of the biochemical reaction, P represents the probability, I represents the probability of the biochemical reaction f(x)≤γ (= 1) indicating function, I f(x)≤γ ＝1、I| f(x)＞γ =0,n denotes the resulting physical location layout x of the component i The number of the cells.
10. The system of claim 7, wherein the routing layout module is configured to calculate the biochemical reaction completion time in a plurality of routing layouts corresponding to each of the component physical location layouts according to a maze algorithm, respectively;
correspondingly, the position layout module is specifically configured to select a minimum biochemical reaction completion time as the biochemical reaction completion time for the physical position layout of the components, and select a routing layout corresponding to the minimum biochemical reaction completion time as a connection scheme for each component in the physical position layout of the components.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510279070.XA CN104881542B (en) | 2015-05-27 | 2015-05-27 | The component method for arranging and system of biochemical reaction detection device on micro chip |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510279070.XA CN104881542B (en) | 2015-05-27 | 2015-05-27 | The component method for arranging and system of biochemical reaction detection device on micro chip |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104881542A CN104881542A (en) | 2015-09-02 |
| CN104881542B true CN104881542B (en) | 2018-05-22 |
Family
ID=53949035
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510279070.XA Expired - Fee Related CN104881542B (en) | 2015-05-27 | 2015-05-27 | The component method for arranging and system of biochemical reaction detection device on micro chip |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104881542B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105844086B (en) * | 2016-03-18 | 2019-02-15 | 中国人民武装警察部队工程大学 | A kind of arragement construction of pressure sensor |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1560771A (en) * | 2004-02-20 | 2005-01-05 | 清华大学 | Integrated circuit module level distributing method based on module deformation and probability local search |
| US6892366B1 (en) * | 2001-08-28 | 2005-05-10 | Cadence Design Systems, Inc. | Method and apparatus for performing extraction using a model trained with Bayesian inference via a Monte Carlo method |
| CN104462729A (en) * | 2014-12-31 | 2015-03-25 | 中国电子科技集团公司第四十七研究所 | Layout method for anti-fuse field-programmable gate array |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9396302B2 (en) * | 2013-10-02 | 2016-07-19 | Utah State University | Global router using graphics processing unit |
-
2015
- 2015-05-27 CN CN201510279070.XA patent/CN104881542B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6892366B1 (en) * | 2001-08-28 | 2005-05-10 | Cadence Design Systems, Inc. | Method and apparatus for performing extraction using a model trained with Bayesian inference via a Monte Carlo method |
| CN1560771A (en) * | 2004-02-20 | 2005-01-05 | 清华大学 | Integrated circuit module level distributing method based on module deformation and probability local search |
| CN104462729A (en) * | 2014-12-31 | 2015-03-25 | 中国电子科技集团公司第四十七研究所 | Layout method for anti-fuse field-programmable gate array |
Non-Patent Citations (6)
| Title |
|---|
| Hierarchical Cross-Entropy Optimization for Fast On-Chip Decap Budgeting;Xueqian Zhao等;《IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTERGRATED CIRCUITS AND SYSTEMS》;20111017;第30卷(第11期);第1610-1620页 * |
| Layout-Aware Solution Preparation for Biochemical Analysis on a Digital Microfluidic Biochip;Sudip Roy等;《International Conference on Vlsi Design》;20110222;第171-176页 * |
| 二维矩形块布局的交叉熵方法实现;吕盛坪等;《北京交通大学学报》;20090415;第33卷(第2期);第39-43页 * |
| 基于交叉熵方法的布局问题求解算法研究;卢长先;《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》;20070615(第06期);第I138-5页 * |
| 数字微流控生物芯片布局的拟人遗传组合算法;杨敬松等;《计算机工程与应用》;20120828;第48卷(第31期);第16-20页 * |
| 数字微流控生物芯片的布局及调度问题研究;杨敬松;《中国博士学位论文全文数据库 信息科技辑》;20090715(第07期);第I135-37页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104881542A (en) | 2015-09-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111260019B (en) | Data processing method, device and equipment of neural network model and storage medium | |
| US20130178968A1 (en) | Methods, Systems, and Products for Conducting Droplet Operations | |
| Liu et al. | Transport or store? Synthesizing flow-based microfluidic biochips using distributed channel storage | |
| US20230152367A1 (en) | System and methods for analyzing and estimating susceptibility of circuits to radiation-induced single-event-effects | |
| Wang et al. | An estimation of distribution algorithm for scheduling problem of flexible manufacturing systems using Petri nets | |
| CN106203622A (en) | Neural network computing device | |
| Liu et al. | Design of optimal Petri-net controllers for a class of flexible manufacturing systems with key resources | |
| Liu et al. | DCSA: Distributed channel-storage architecture for flow-based microfluidic biochips | |
| Chakraborty et al. | Efficient two phase heuristic routing technique for digital microfluidic biochip | |
| Zhu et al. | Multi-channel and fault-tolerant control multiplexing for flow-based microfluidic biochips | |
| Zhu et al. | Multicontrol: Advanced control-logic synthesis for flow-based microfluidic biochips | |
| Nedjah et al. | Preference-based multi-objective evolutionary algorithms for power-aware application mapping on NoC platforms | |
| US20230186007A1 (en) | System and method for optimizing integrated circuit layout based on neural network | |
| Liu et al. | Design automation for continuous-flow microfluidic biochips: A comprehensive review | |
| CN104881542B (en) | The component method for arranging and system of biochemical reaction detection device on micro chip | |
| Farooq et al. | Efficient FPGA routing using reinforcement learning | |
| CN111141920B (en) | On-line fault detection method of digital microfluidic biochip based on reinforcement learning | |
| Ibrahim et al. | Integrated and real-time quantitative analysis using cyberphysical digital-microfluidic biochips | |
| Khare et al. | Task mapping and flow priority assignment of real-time industrial applications for network-on-chip based design | |
| Keszocze et al. | Exact design of digital microfluidic biochips | |
| Rajesh et al. | Invasive weed optimization based scheduling for digital microfluidic biochip operations | |
| TWI767303B (en) | Computer-implemented method of propagation latency reduction in neural network | |
| Tashev et al. | A numerical study of the upper bound of the throughput of a crossbar switch utilizing MiMa-algorithm | |
| Ji et al. | More effective randomly-designed microfluidics | |
| CN116205023B (en) | Design method, device, equipment and storage medium of intelligent robot laboratory |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20171229 Address after: 100045, 1, 16 Hebei street, three li, Beijing, Xicheng District Applicant after: Midtown characteristic town planning Research Institute Co Ltd Address before: 100083 Beijing City, Haidian District Zhongguancun Road No. 18 smartfortune International Building A-605 Applicant before: Xie Jun Applicant before: Hu Shiyan |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180522 Termination date: 20190527 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |