CN104346399A - Method for analyzing basic data of Chinese herbal compound based on PK-PD (Pharmacokinetics-Pharmacodynamics) - Google Patents
Method for analyzing basic data of Chinese herbal compound based on PK-PD (Pharmacokinetics-Pharmacodynamics) Download PDFInfo
- Publication number
- CN104346399A CN104346399A CN201310339722.5A CN201310339722A CN104346399A CN 104346399 A CN104346399 A CN 104346399A CN 201310339722 A CN201310339722 A CN 201310339722A CN 104346399 A CN104346399 A CN 104346399A
- Authority
- CN
- China
- Prior art keywords
- index
- data
- time point
- chinese medicine
- material base
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/40—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for data related to laboratory analysis, e.g. patient specimen analysis
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
Landscapes
- Public Health (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Data Mining & Analysis (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
A method for analyzing the basic data of a Chinese herbal compound based on PK-PD (Pharmacokinetics-Pharmacodynamics) comprises the following steps: 1) conducting steady transformation on original PK and PD data obtained by experiments; 2) drawing a virtual observation sample at random based on the empirical distribution function of the PK or PD data of each time point by a re-sampling method based on the overall observation numerical value of the sample, forming a virtual sample orbit through the virtual observation of multiple time points, and simulating the PK or PD measurement process; 3) calculating the accumulative effect of each time point, respectively calculating the relationships between selected PK indexes and PD indexes, and analyzing differences among different groups; 4), calculating the variation rates of the PD indexes of each virtual sample orbit about each herbal ingredient at each time point, and obtaining the action effect of the PK indexes on the PD indexes through the density evolution process of the variation rates. The relationships between the ingredients and the PD indexes are found through the quantitative analysis of the variation relationships between the PK data and the PD data.
Description
Technical field
The present invention relates to data processing field, more specifically, relate to the Chinese medicine compound prescription material base data analysing method based on PK-PD.
Background technology
Chinese medicine compound prescription is a complication system be made up of multiple flavour of a drug, and because of the therapeutic action that it is special to disease, the material base research relevant to effect becomes the interest of world scholar always.
In discussion at present for Chinese medicine compound prescription material base, PK-PD joint study is the method wherein meeting theory of traditional Chinese medical science the most.
PK, i.e. pharmacokinetics (pharmacokinetics), the principle of finger power and mathematic(al) mode, quantitative description medicine enters the rule of blood concentration dynamic change in time in absorption in body, distribution, metabolism and discharge process, illustrate the effect of body to medicine, namely PK tables of data understands the relation in " time m-constituent concentration ".PD, i.e. pharmacodynamics (pharmacodynamics, PD), the dynamic process that drugs effect changes along with administration time, illustrates the effect of medicine to body, namely PD tables of data understand " time m-effect (index) relation.Within the quite a long time, PK and PD is two independently subjects, and along with going deep into of research, people recognize gradually, and isolated research PK or PD, ignore contact between the two, the information obtained is incomplete.
Propose PK and PD models coupling to get up from the sixties in 20th century, existing more than 50 year so far, PK-PD model developed into the system of a set of science, but above-mentioned theory and model are based under the more clearly prerequisite of mechanism of drug action.Namely drug ingedient structure is known, and its pharmacodynamics effect produced in vivo is caused by this composition, and both relations are clear and definite.Inquire into the relation between " Concentration-time-effect " three by modeling, can comparatively accurately and comprehensively predict and describe the time dependent rule of effect of medicine under certain dosage and dosage regimen, instruct its clinical practice.
But the PK-PD of Chinese medicine compound prescription studies then different from above-mentioned, its relation wants complicated many.This is because Chinese medicine compound prescription plays the feature that curative effect has multicomponent, too many levels, Mutiple Targets, a certain chemical composition may act on multiple PD index, a certain PD index also may be contributed by multiple composition, therefore cannot continue to use the discussion that original theory and model carry out Chinese medicine compound prescription material base.
In addition, for Chinese medicine compound prescription PK data, usually adopt the instruments such as HPLC-UV or HPLC-MS, collect blood or the tissue of different time points, carried out the content that instrumental analysis goes out sample Chinese traditional medicine composition.For PD data, be the feature according to selected drug therapy disease, determine PD index, carried out the mensuration of PD data by pharmacological method.Due to complicated component in Chinese medicine, Pharmacodynamical mechanism is unclear, therefore will obtain the PK data of multiple composition, and the PD data of multiple drug effect.Therefore, the Data Analysis Services problem how solving multi-input multi-output system needs now the problem of solution badly.In addition, in data handling procedure, also there will be the large and nonsynchronous problem of PK/PD data time of experimental data body difference, and the false-positive problem of experimental data.
Summary of the invention
For the problems referred to above, the present invention proposes a kind of Chinese medicine compound prescription material base data analysing method based on PK-PD, comprising: 1) original PK and the PD data that experiment obtains steadily and surely are converted; 2) adopt based on the overall repeat replication of the observation value of sample, empirical distribution function based on PK or the PD data of each time point randomly draws an observation sample, thus the observation data of multiple time point just forms a path, simulates pharmacokinetics or the pharmacodynamics measuring process of a subjects with this; 3) calculate the cumulative effect of each time point, and calculate the correlationship between selected PK index and PD index respectively, analyze the difference between different group; 4) calculate the PD index of each path about the rate of change of each drug ingedient at each time point, and then obtain the action effect of pharmacokinetic indicator for pharmacodynamic index by the density evolution process of rate of change.
The research greatest problem of Chinese medicine compound prescription does not know what material base is, do not know which composition onset.The data analysis processing method based on PK-PD that the present invention sets up, that mechanism Chinese medicine compound prescription played a role in vivo regards Black smoker as, multicomponent pharmacokinetics PK data are the input of Black smoker, and the pharmacodynamics PD data of the corresponding multi objective produced are the output of Black smoker.By quantitative description and the variation relation analyzing PK data and PD data, therefrom find and find which composition has contribution to which PD index, and its percentage contribution has how many, namely obtain the relation of " composition-drug effect (index) ".The present invention is that the research of Chinese medicine compound prescription material base provides a new thinking and countermeasure.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of an embodiment of method of the present invention;
Fig. 2 is 100 analog results of three PD indexs at 6 time points and the sectional view of cumulative effect thereof;
Fig. 3 A is the impact of the 3rd pharmacokinetic indicator for the 1st pharmacodynamic index;
Fig. 3 B is the impact of the 3rd pharmacokinetic indicator for the 2nd pharmacodynamic index;
Fig. 4 A-4C is control group and the ginseng group PD index PR trend map of eliminating individual difference;
Fig. 5 A is the cumulative effect curve of each processed group in arterial contraction forcing up the targets;
Fig. 5 B is the cumulative effect curve of each processed group in cardiac output index;
Fig. 5 C is the cumulative effect curve of each processed group in QRS interval index.
Embodiment
First embodiment:
According to first embodiment of the invention.Method of the present invention is for pharmacokinetics PK data, by the relation of quantitative description and analysis ingredient and drug effect, therefrom find and find which composition has contribution to which PD index, and its percentage contribution has how many, and then the pharmacodynamics obtaining multi objective produces the data of respective change.Method of the present invention comprises the steps:
1, to original PK and the PD data that experiment obtains, calculate the empirical distribution function of each point in time measurement data, thus measurement data is steadily and surely converted.
Original PK and the PD data that experiment obtains fluctuation is in time general larger, and extremely unstable, trend term is also not obvious, and the dimension of data differs.Therefore, need steadily and surely to convert PK and PD data.
Sane conversion can adopt empirical distribution function, and the value of empirical distribution function is generally the division operation of two integers, for each PK or PD measurement index X, by the measurement data { x of individual for all tests i and different time points j for this index
i,jregard a complete data set as, total sample size of this set is N, then constructs the empirical distribution function of this data acquisition:
Wherein, I is indicator function, (x
i,j< t) value is 1 when meeting, otherwise be 0.
Like this, each observed reading x of measurement index X
i,j, empirical distribution function F (t) about this entirety must have a functional value F (x
i,j), then this functional value F (x
i,j) just as observational variable x
i,jsane conversion numerical value.Such as, for certain observed reading x
i,j, other similar observed reading might as well be recorded as x
m,n, then all x
m,nwith x
i,jcarry out size relatively and obtain the value of a series of 0 or 1, wherein 1 all 0, in 1 sequence, shared ratio is exactly x
i,jtransformed value.Can find out, through such conversion process, the transformed value of each data, all in (0,1) interior value, remains the relative size relation of each numerical value observation similar with other simultaneously.
The present invention is by the sane conversion based on overall distribution, eliminate the dimension of each data, ensure that the relative size relation of data, make the data of those abnormal values obtain more appropriate suppression simultaneously, and then various quantitative analysis can be carried out by the quantitative value after conversion
More advantageously, can carry out correction for continuity to empirical distribution function by linear interpolation method, namely the distribution function value of each point can be determined by the linear interpolation of the functional value of two points faced recently with it.When data volume is smaller, the method can improve operational precision, then affects little when data volume is larger.
More advantageously, can also to the descriptive statistic of each measurement index.The basic statistics amount such as maximal value, minimum value, median of descriptive statistic basic index certificate and the figure of data describe, and these descriptive statistics can provide visual information.
2, adopt based on the overall repeat replication of the observation value of sample, empirical distribution function based on PK or the PD measurement data of each time point randomly draws an observation sample, the observation data of multiple like this time point just forms a path, and then simulates pharmacokinetics or the pharmacodynamics measuring process of a subjects with this.
Preferably, the described repeat replication overall based on the observation value of sample is Bootstrap repeat replication.The reliability of Bootstrap repeat replication is ensured by the stability of empirical distribution function, and each sampling all obtains the virtual observation that is different from original observation, thus can make up the less weakness of sample size.This feature of Bootstrap repeat replication is obvious compared with the repeat replication of the overall based on the observation value of sample of other type.
This method can make up the deficiency of data and contribute to weakening the impact that causes more greatly of data fluctuations: consider that PK and PD data are observed at different individualities, individual difference is comparatively large and data volume is less, but the empirical distribution function of observation data meets stability usually, and insensitive to the observation of exception, it is reliable that the empirical distribution function that therefore the present invention is based on observation carries out Bootstrap double sampling to construct the virtual observation of each time point.
This step 2 comprises:
(2-1) all data of each time point in each factor (drug ingedient or drug effect) are regarded as one totally, and utilize these data configuration empirical distribution functions,
(2-2) from corresponding overall of these time points, randomly draw a virtual sample respectively, thus form a virtual observation.Method of randomly drawing is such as Bootstrap repeat replication.
(2-3) repeat the step of (2-1)-(2-2), carry out certain number of times (such as 200 times), be so just equivalent to carry out 200 tests.
The method by from preset time data distribution function extract abundant virtual sample, be equivalent to add experiment sample, thus careful quantitative analysis (such as probability density evolution method) can be compared accordingly, improve the reliability of analysis.
3, calculate the cumulative effect of each time point, and calculate the correlationship between user-selected PK index and PD index respectively, analyze the difference of drug effect between different group.
In addition, can also to through time PD index (the PD data of multiple time point can be obtained) and terminal PD index (the PD data of a time point can only be obtained) between carry out correlation analysis.If this endpoint is goldstandard, then utilize different PD index about the correlativity of this goldstandard, the alternative Composition analyzed index of the goldstandard that may there is unfavorable factor in reality can also be inferred.
Described step 3 comprises the steps:
(3-1) cumulative effect of each time point is calculated
PD index is very large in the value fluctuation of each time point, directly considers that the effect of each time point is insecure, and the Changing Pattern that the value distribution of pharmacokinetics pharmacodynamic index is only discussed from the angle of distribution is only reliable.
Therefore, the present invention, according to statistical ultimate principle, supposes that medicine is invalid, then different time points should be random fluctuation about the accumulative effect of the difference of positive control, can be regarded as the Brownian movement that is average with 0.Based on such hypothesis, definition cumulative effect index:
Here X
ithe difference of observed reading about reference point of i-th time point.Such as, Fig. 2 shows three PD indexs in 100 analog results of 6 time points and cumulative effect thereof.As can be seen from Figure 2, the cumulative effect of different observation index meets certain level in the value of each time point, has substantially had relatively clear and definite value variation tendency, can be further analyzed.
(3-2) Pearson correlation coefficient between PK index and PD index is calculated, concrete grammar is: for selected PK index and PD index, first obtain the individual all combination of two of test, then PK index and PD index are calculated respectively at each time point t for each combination
idifference, all these differences form a mutually corresponding 2-D data sequence, then namely the Pearson correlation coefficient of this sequence can measure the correlativity between PK index and PD index.When PD index only has the value of a time point, the method for calculating is similar, and only PK index is no longer calculate the corresponding difference of each time point, but calculate PK through time curve under the total area in the difference of different tests individuality combination.
(3-3) correlativity of different PD index about drug effect gold index is calculated, if certain PD index and drug effect gold index have higher correlativity, just can consider that substituting some by this PD index has nocuity or the golden index of the drug effect that not easily detects to pass judgment on the effect of medicine so in practice.
Specifically, if the result for the treatment of of medicine can be decided by the effect testing k the drug ingedient measured, and related coefficient is the index of a reliable explanation drug ingedient effect, so k drug ingedient just can be regarded as a proper vector T of tolerance effect of drugs about the related coefficient of golden index
g.Similar, each auxiliary PD index also can obtain one with the proper vector of k related coefficient composition.Obviously, if certain auxiliary characteristics A characteristic of correspondence vector T
awith the proper vector T of golden index
grelatively more consistent, i.e. T
aand T
gbetween related coefficient have larger value, then mean this auxiliary characteristics likely substitute golden index to judge the effect of medicine.
4, the PD index of each virtual sample track is calculated about the density Estimation of each drug ingedient at each time point, and then by probability density function along with the evolution trend of time obtains the action effect of pharmacokinetic indicator for pharmacodynamic index.
More specifically, Cubic Spline Method is utilized to calculate the PD index of each virtual rail about the numerical solution of each drug ingedient at the partial derivative of each time point.Then utilize kernel density estimation method to obtain the density Estimation of each time point PD index about the partial derivative of each drug ingedient, and then by the evolution trend of probability density function along with the time, pharmacokinetic indicator is judged for onset time of pharmacodynamic index, duration action time and the effect such as promotion or suppression.
In sum, the cumulative effect that the present invention is based on pharmacokinetics pharmacodynamic index calculates the rate of change of pharmacodynamic index about pharmacokinetic indicator, and is described by the relation of density evolution process to pharmacokinetics pharmacodynamic index of rate of change.
The observation data of pharmacokinetic indicator and pharmacodynamic index has larger variation, the present invention is on the basis of virtual rail, estimate that pharmacodynamic index is at the rate of change of each time point about pharmacokinetic indicator by Cubic Spline Method, the value of a series of rate of change can be obtained like this at each drug ingedient of each time point and PD index for each virtual rail, then calculate the probability density value of each this rate of change of time point, judge the influence degree of specific pharmacokinetic indicator for pharmacodynamic index accordingly.
In general, if certain drug ingedient does not have obvious effect for drug effect, the probability density of partial derivative will be concentrated near 0 value, otherwise will move to other on the occasion of or negative value place.In addition, for certain drug ingedient, As time goes on, rate of change generally can be got back to again near 0 value, usually mean that the metabolic process of this drug ingedient terminates, thus just can infer the onset time of certain composition and the length of duration according to the transformation period of figure display.
Such as, referring to the density evolution figure of certain drug ingedient shown in Fig. 3 A, 3B about PD index, wherein Fig. 3 A is the impact of the 3rd pharmacokinetic indicator for the 1st pharmacodynamic index.Fig. 3 B is the impact of the 3rd pharmacokinetic indicator for the 2nd pharmacodynamic index.As can be seen from Fig. 3 A and 3B, same pharmacokinetic indicator often has similar shape for different pharmacodynamic indexs, illustrates that this composition has relatively clear and definite mechanism of action, otherwise just may there is stronger interaction with other composition.Can judge from Fig. 3 A and 3B, third and fourth drug ingedient is for the obvious effect of result for the treatment of.From figure, this composition moves larger about the probability density value of pharmacodynamics near 0 value to negative value after certain time, illustrate that this composition is a kind of inhibiting effect to current PD index, this composition is likely a kind of short-term effect in addition, because display density curve continues for some time the movement near 0 value again of rear density in figure.
Second embodiment:
Data are steadily and surely converted, the abnormal value of data can be suppressed, but the difference between individuality cannot be eliminated.According to a further aspect in the invention, in order to improve precision of analysis, the large and nonsynchronous problem of PK/PD data time of experimental data individual difference also to be considered.Therefore, the present invention proposes second embodiment, the method comprising the steps of:
1, steadily and surely convert (after the step 1) of first embodiment to the observation data of reality, the present invention adopts base-line method to carry out the further process of experimental data, using the observed reading of ill original state as baseline criteria, calculate the difference of observation relative to baseline of each time point, using the apparent curve of difference as the PD index changed along with time point, (control group of indication refers in pharmaceutical research such control group Trendline herein, except not giving Experimental agents, the operational group of the equivalent formal experiment of other operation.The object of control group setting is to eliminate in experimental study except medicine other extraneous factor to the impact of animal used as test PD index.) mean value curve just can as the reference curve of this PD index of different pharmaceutical processed group, thus weaken individual difference for the impact analyzed.
The variation testing individual original state value due to each is very large, even can not ensure the ordering relationship between disease-free state and ill original state.Therefore, the present invention selects Baseline Methods, using the primary data of just having fallen ill as baseline criteria, considers the difference of PD index about the observed reading of ill original state of different time points, using the apparent curve of this difference along with time point change as drug effect.Such as, for the data of certain medicine after above-mentioned processing procedure, can find out and really there is larger difference between control group and administration group, there is obvious curative effect (as shown in figs. 4 a-4 c) in medicine, therefore can apply the association analysis research that these data carry out next step PK-PD on animal model.
2, for the difference illustratively variable of often kind of composition by this component content in Different Individual serum under each identical time point, simultaneously by the difference responsively variable of the PD index of corresponding individuality, thus change the impact between correlation analysis pharmacokinetics between pharmacodynamic change and pharmacodynamics by medicament contg.This not only weakens the lag-effect of the drug ingedient of different time points, and indirectly can increase sample size, improves the fiduciary level analyzed.
The present invention is based on the correlativity of the difference of each composition of different pharmaceutical and both differential analyses of PD index, the method can weaken drug ingedient and the PD index lag-effect about the time, and achieves the linearization of correlationship to a certain extent.Notice that the onset time of medicine likely exists lag-effect, so the change of the medicament contg in measured serum directly can not correspond to the change of PD index, the drug ingedient that blood concentration is larger compares the variable quantity of the PD index that the less drug ingedient of blood concentration causes usually also can be larger, and Changing Pattern meets certain relation, therefore the present invention is for the difference illustratively variable of often kind of composition by this component content in Different Individual serum under each identical time point, simultaneously by the difference responsively variable of the PD index of corresponding individuality, thus change the impact between correlation analysis pharmacokinetics between pharmacodynamic change and pharmacodynamics by medicament contg.Suppose Y
itthe measurement numerical value of individual i in certain drug effect of t, X
itbe then the blood concentration of individual i at certain composition of t, then can define:
ΔY
t＝Y
it-Y
jt;ΔX
t＝X
it-X
jt
Like this for often kind of composition, altogether can obtain at each time t
individual combination, not only weakens the lag-effect of the drug ingedient of different time points, and indirectly can increase sample size, improves the fiduciary level analyzed.
3rd embodiment:
Consider interaction to be there is between each index, and the correlationship between two two indexes can not by expressing this effect completely, even may there is false-positive problem, in order to improve the accuracy of analysis further, on the basis of second embodiment, the invention allows for the 3rd embodiment.
In the method, the observation data of reality steadily and surely to be converted and after eliminating individual difference, notice that experimental group is not likely 1 group, experimentally designing may be also k group, the present invention is based on the relation of distribution distance to different group PD index and carries out qualitative analysis.
Consider the variability that PD Indexes Comparison is large, therefore drug effect mean value is not adopted, but utilize the PD index of different group to judge the similarity degree between each group at each time point about the distribution distance of control group data, and then infer the effect of different group medicine.This curve can judge that different group is for the whole structure in certain PD index on the whole, and therefore acquired results has certain globality.
Qualitative analysis is carried out based on the relation of distribution distance to different group PD index, might as well suppose observation index have s effective time point, the method utilizes the PD index of different group to judge the similarity degree between each group at each time point about the distribution distance of control group data, and then infer the effect of different group medicine: because group each in test group and control group has n discrete value in the observation data of time t, might as well be assumed to be: { x
t1, x
t2..., x
tn, { y
t1, y
t2..., y
tn, so the trend function of i-th test group and control group is expressed as:
Here I (x
ti< u) be mathematically called indicator function, be meant to as observed reading x
tiduring < u, value is 1, otherwise is 0.Due to trend function
(u) and G
tu () is step function, so the difference of two trend functions is only at the observation data { x of test group and control group
t1, x
t2..., x
tnand { y
t1, y
t2..., y
tnobservation station place just may have different value.Therefore, need two groups of data mixing during Computation distribution distance, total 2n branch, might as well be expressed as: { z
t1, z
t2..., z
t2n, then distribution distance can be expressed as:
Then by each group i each time point t's
connect into Trendline, then can compare the relation of each group.
As shown in figures 5a-5c, curve 1 represents control group, and curve 2 represents ginseng group, and curve 3 represents corydalis tuber group, and curve 4 represents red sage root group, and asterisk line (★) then represents compound group.Transverse axis is time point, and the longitudinal axis is then through the index amount E of the sign drug effect of conversion
t.
As can be seen from Figure 5A, the effect of compound group that the red sage root group that represents of curve 4 and asterisk line represent presses relatively at arterial contraction.As can be seen from Figure 5B, the effect of compound group that represents of the red sage root group that represents of curve 4 and asterisk line in the indexs such as cardiac output also relatively.But Fig. 5 C illustrates, then there is larger separation in both to the effect of the compound group that the red sage root group that curve 4 represents represents with asterisk line in QRS interval index.This has different effects with regard to the different composition of both promptings or there is interaction on QRS, thus contributes to the effect analyzing heterogeneity.
Above-described embodiment is typical embodiment of the present invention; but the present invention is not restricted to the described embodiments; change, the modification done under other any does not deviate from Spirit Essence of the present invention and principle, substitute, combine, simplify; all should be the substitute mode of equivalence, be included within protection scope of the present invention.
Claims (12)
1., based on a Chinese medicine compound prescription material base data analysing method of PK-PD, it is characterized in that, comprising:
1) original PK and the PD data that experiment obtains steadily and surely are converted;
2) adopt based on the overall repeat replication of the observation value of sample, empirical distribution function based on PK or the PD data of each time point randomly draws an observation sample, thus the observation data of multiple time point just forms a path, simulates pharmacokinetics or the pharmacodynamics measuring process of a subjects with this;
3) calculate the cumulative effect of each time point, and calculate the correlationship between selected PK index and PD index respectively, analyze the difference between different group;
4) calculate the PD index of each path about the rate of change of each drug ingedient at each time point, and then obtain the action effect of PD index for PK index by the density evolution process of rate of change.
2. the Chinese medicine compound prescription material base data analysing method based on PK-PD according to claim 1, is characterized in that, in described step 1), by calculating the empirical distribution function of each point in time measurement data, steadily and surely converts data.
3. the Chinese medicine compound prescription material base data analysing method based on PK-PD according to claim 2, is characterized in that, also comprise:
By linear interpolation method, namely the distribution function value of each point is determined by the linear interpolation of the functional value of two points faced recently with it, carries out correction for continuity to empirical distribution function.
4. the Chinese medicine compound prescription material base data analysing method based on PK-PD according to claim 1, is characterized in that, in step 2) in, the described repeat replication overall based on the observation value of sample is Bootstrap repeat replication.
5. the Chinese medicine compound prescription material base data analysing method based on PK-PD according to claim 4, is characterized in that, described step 2) comprising:
2-1) all data of time point each in PK or PD data are regarded as one totally, and utilize these data configuration empirical distribution functions,
2-2) from corresponding overall of these time points, randomly draw a sample respectively, thus form and once observe;
2-3) repeat step 2-1)-2-2) to pre-determined number.
6. the Chinese medicine compound prescription material base data analysing method based on PK-PD according to claim 1, it is characterized in that, described step 3) comprises:
3-1) calculate the cumulative effect of each time point;
3-2) calculate the Pearson correlation coefficient between PK index and PD index;
3-3) calculate the correlativity of different PD index about drug effect gold index.
7. the Chinese medicine compound prescription material base data analysing method based on PK-PD according to claim 6, it is characterized in that, described step 3) also comprises:
3-4) analyze through time PD index and terminal PD index between correlativity;
If 3-5) this endpoint is goldstandard, then utilize different PD index about the correlativity of this goldstandard, infer the alternative Composition analyzed index that may there is the goldstandard of unfavorable factor in reality.
8. the Chinese medicine compound prescription material base data analysing method based on PK-PD according to claim 1, it is characterized in that, described step 4) comprises:
4-1), Cubic Spline Method is utilized to calculate the PD index of each track about the numerical solution of each drug ingedient at the partial derivative of each time point;
Kernel density estimation method 4-2) is utilized to obtain the density Estimation of each time point PD index about the partial derivative of each drug ingedient;
4-3) by probability density function along with the evolution trend of time obtains the PD index action effect for PK index.
9. the Chinese medicine compound prescription material base data analysing method based on PK-PD according to claim 1, is characterized in that, also comprise:
5) based on the correlativity of the difference of each composition of different pharmaceutical and both differential analyses of PD index.
10. the Chinese medicine compound prescription material base data analysing method based on PK-PD according to claim 9, it is characterized in that, described step 5) comprises:
5-1) using the observed reading of ill original state as baseline criteria, calculate the difference of observation relative to baseline of each time point;
5-2) using the apparent curve of difference as the PD index changed along with time point;
5-3) using the reference curve of the mean value curve of control group Trendline as this PD index of different pharmaceutical processed group;
5-4) for the difference illustratively variable of often kind of composition by this component content in Different Individual under each identical time point, simultaneously by the difference responsively variable of the PD index of corresponding individuality, change the relation between correlation analysis pharmacokinetics between pharmacodynamic change and pharmacodynamics by medicament contg.
The 11. Chinese medicine compound prescription material base data analysing methods based on PK-PD according to claim 1, is characterized in that, also comprise:
6) utilize the PD index of different group to judge the similarity degree between each group at each time point about the distribution distance of control group data, and then infer the effect of different group medicine.
The 12. Chinese medicine compound prescription material base data analysing methods based on PK-PD according to claim 11, it is characterized in that, in described step 6), for some effective time of observation index, the PD index of different group is utilized to judge the similarity degree between each group at each time point about the distribution distance of control group data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310339722.5A CN104346399B (en) | 2013-08-06 | 2013-08-06 | Method for analyzing basic data of Chinese herbal compound based on PK-PD (Pharmacokinetics-Pharmacodynamics) |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310339722.5A CN104346399B (en) | 2013-08-06 | 2013-08-06 | Method for analyzing basic data of Chinese herbal compound based on PK-PD (Pharmacokinetics-Pharmacodynamics) |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104346399A true CN104346399A (en) | 2015-02-11 |
| CN104346399B CN104346399B (en) | 2017-02-15 |
Family
ID=52502012
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310339722.5A Active CN104346399B (en) | 2013-08-06 | 2013-08-06 | Method for analyzing basic data of Chinese herbal compound based on PK-PD (Pharmacokinetics-Pharmacodynamics) |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104346399B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104897791A (en) * | 2015-03-25 | 2015-09-09 | 中国中医科学院中药研究所 | Simulation and prediction method for pharmacokinetic curve of traditional Chinese medicine |
| CN107064480A (en) * | 2016-08-12 | 2017-08-18 | 株洲千金药业股份有限公司 | Chinese medicinal ingredients act on azithromycin baseline lifting evaluation method and evaluate the application of gynaecology's prescriptions worth thousand gold |
| CN108628889A (en) * | 2017-03-21 | 2018-10-09 | 北京京东尚科信息技术有限公司 | Sampling of data mthods, systems and devices based on timeslice |
| CN112837756A (en) * | 2021-01-20 | 2021-05-25 | 中国人民解放军军事科学院军事医学研究院 | Method for deducing exposure characteristics of compounds in plasma from urine excretion curve |
| CN117457114A (en) * | 2023-12-25 | 2024-01-26 | 大连锦辉盛世科技有限公司 | Intelligent storage management method for western medicine pharmaceutical composition data |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101615222A (en) * | 2008-06-23 | 2009-12-30 | 中国医学科学院放射医学研究所 | A kind of Chinese prescription designing technique based on the Chinese medicine effective component group |
| CN101780112A (en) * | 2009-01-21 | 2010-07-21 | 辽宁省中医药研究院 | Screening method of active ingredients of traditional Chinese medicine |
-
2013
- 2013-08-06 CN CN201310339722.5A patent/CN104346399B/en active Active
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104897791A (en) * | 2015-03-25 | 2015-09-09 | 中国中医科学院中药研究所 | Simulation and prediction method for pharmacokinetic curve of traditional Chinese medicine |
| CN104897791B (en) * | 2015-03-25 | 2017-01-04 | 中国中医科学院中药研究所 | A kind of Chinese medicines is for dynamic (dynamical) curve simulation and Forecasting Methodology |
| CN107064480A (en) * | 2016-08-12 | 2017-08-18 | 株洲千金药业股份有限公司 | Chinese medicinal ingredients act on azithromycin baseline lifting evaluation method and evaluate the application of gynaecology's prescriptions worth thousand gold |
| CN107064480B (en) * | 2016-08-12 | 2019-05-03 | 株洲千金药业股份有限公司 | Chinese medicinal ingredients promote evaluation method to azithromycin effect baseline and evaluate the application of gynaecology's prescriptions worth thousand gold |
| CN108628889A (en) * | 2017-03-21 | 2018-10-09 | 北京京东尚科信息技术有限公司 | Sampling of data mthods, systems and devices based on timeslice |
| CN108628889B (en) * | 2017-03-21 | 2021-05-25 | 北京京东尚科信息技术有限公司 | Time slice-based data sampling method, system and device |
| CN112837756A (en) * | 2021-01-20 | 2021-05-25 | 中国人民解放军军事科学院军事医学研究院 | Method for deducing exposure characteristics of compounds in plasma from urine excretion curve |
| CN117457114A (en) * | 2023-12-25 | 2024-01-26 | 大连锦辉盛世科技有限公司 | Intelligent storage management method for western medicine pharmaceutical composition data |
| CN117457114B (en) * | 2023-12-25 | 2024-03-12 | 大连锦辉盛世科技有限公司 | Intelligent storage management method for western medicine pharmaceutical composition data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104346399B (en) | 2017-02-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bland et al. | Comparing two methods of clinical measurement: a personal history | |
| Bland et al. | Statistical methods for assessing agreement between two methods of clinical measurement | |
| CN104346399B (en) | Method for analyzing basic data of Chinese herbal compound based on PK-PD (Pharmacokinetics-Pharmacodynamics) | |
| Zhang et al. | Five algorithms that calculate cardiac output from the arterial waveform: a comparison with Doppler ultrasound | |
| Park et al. | Transthoracic echocardiography versus computed tomography for ascending aortic measurements in patients with bicuspid aortic valve | |
| Zhang et al. | 4D flow MRI pressure estimation using velocity measurement-error-based weighted least-squares | |
| CN103020958A (en) | Vessel automatic matching method based on curvature scale space | |
| Bates et al. | A nonstatistical approach to estimating confidence intervals about model parameters: application to respiratory mechanics | |
| EP3202319B1 (en) | Method for detecting position of signal source in living body, and device for detecting position of signal source in living body | |
| CN100453038C (en) | Method and apparatus for analyzing organism reaction waveform information, and diagnosing apparatus | |
| Kang et al. | A sample size computation method for non‐linear mixed effects models with applications to pharmacokinetics models | |
| CN107920762A (en) | The devices and methods therefor of systolic pressure is calculated using Pulse transit time | |
| Karavelioglu et al. | Gender differences in hand grip strength of the child athletes by using absolute, ratio and allometric scaling methods | |
| Yang et al. | Harmonic skeleton guided evaluation of stenoses in human coronary arteries | |
| Zuber et al. | Effect of reversible hypoperfusion on left ventricular volumes measured with gated SPECT at rest and after adenosine infusion | |
| CN105930646A (en) | Data processing system and method for evaluating heart aging degree | |
| Munro et al. | An analysis of the errors in certain anthropometric measurements | |
| Wang | A comparison of alternative estimation methods in confirmatory factor analyses of the General Health Questionnaire across four groups of australian immigrants | |
| KR102565228B1 (en) | Apparatus and method for measuring cardiopulmonary functions | |
| Aldroubi et al. | Optimal sampling for the determination of calcium metabolism | |
| Tiesinga et al. | Sensitivity, specificity, and usefulness of the Dutch Fatigue Scale | |
| Wang et al. | Impact of regular physical activity on patterns of age-related changes in photoplethysmogram appearance obtained at fingertip | |
| RU2648879C1 (en) | Method for defining the surface area of the human body | |
| Gaydosh et al. | Biological Age in Taiwan: Measurement and Prediction | |
| Gardner et al. | Pressure to perform: Is cardiac output estimation from arterial waveforms good enough for routine use? |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |