CN104360001A - Discriminating method for heroin origins - Google Patents

Abstract

The invention provides a discriminating method for heroin origins based on neutral or acidic impurities and partial least-squares discriminant analysis (PLS-DA), which is suitable for discriminating affiliation of heroin in two origins: southeast Asia and southwest Asia. The method comprises the following steps: firstly, putting forward a discriminating function of PLS-DA of the heroin origins; detecting peak area numerical values of independent variables of 25 neutral or acidic impurities in a heroin sample, and substituting the peak area numerical values into the PLS-DA discriminating function; calculating the numerical value t of the obtained discriminating function; when t is greater than 0 and less than 6, judging that the heroin origin is southeast Asia; and when t is greater than -6 and less than 0, judging that the heroin origin is southwest Asia.

Description

A kind of heroin source place method of discrimination

Technical field

The present invention relates to discrimination technology field, heroin source place, particularly a kind of based on neutral or acid impurities and partial least square method techniques of discriminant analysis (PLS-DA) heroin source place method of discrimination.

Background technology

Heroin is as one of maximum drugs of the domestic the most common and amount of sucking at present, and very big to social danger, therefore heroin source more causes the concern of people.The discriminatory analysis in heroin source place will be conducive to analyzing the immigration channel of heroin and transport goods for sale path, can for the Strategic Evaluation of the whole nation or a certain regional heroin poison feelings and for the department that bans taking addictive drugs policy making, information is studied and judged provides Data support.

Heroin, is also called diacetylmorphine, is a kind of semi-synthetic drugs, is obtained after acetylation by the main alkaloid morphine in opium.Heroin also generates a large amount of neutrality or acid impurities in illegal process, and these neutral or acid impurities are that the alkaloid such as morphine, codeine, papaverine, thebaine, narcotine in acetic anhydride and opium reacts and generates.The kind of these impurity contained in heroin and quantity, directly reflect the processing technology of heroin, some degree also can reflect the place of production of opium.Utilize these distinctive neutrality or the acid impurities in heroin, the discriminant classification on heroin separate sources ground can be carried out.Owing to being subject to rainfall amount, the impact of temperature on average and soil types, the area of establishing in large scale creamcups mainly contains 4 large regions in the world at present: Southeast Asia, South-West Asia, South America and Mexico.Usually the creamcups of this real estate is all processed into heroin in this locality.

Existing about adopting neutrality or acid impurities to carry out in the document of heroin source place discriminant classification, all do not adopt any chemometrics method, the kind according to mainly or acid impurities neutral according to feature contained in heroin sample differentiated and relative content, so differentiate that process is comparatively complicated, rely on experience more, result of determination is larger by the impact of human factor.

Summary of the invention

The object of the invention is proposition a kind of based on neutral or acid impurities and PLS-DA heroin source place method of discrimination, the method only needs to know the peak area numerical value of 25 kinds of neutrality or acid impurities in heroin, substitute into the present invention discriminant function is proposed, through simple computation can realize heroin " Southeast Asia " and " South-West Asia " two source place quick, effectively distinguish.

A kind of based on heroin source place method of discrimination that is neutral in heroin or acid impurities, measured by centering or acid impurities, determine the source of heroin, it is characterized in that: detected the peak area numerical value obtaining 25 kinds of neutrality or acid impurities compound in heroin sample by Ultra Performance Liquid Chromatography-series connection quadrupole rod time-of-flight mass spectrometry (TOFMS) (UPLC-Q-TOF) method, process is carried out to above-mentioned peak area numerical value and obtains threshold values of classifying, determine Southeast Asia, heroin source place, the distributed area of South-West Asia, calculate the numerical value of unknown heroin sample, Southeast Asia or South-West Asia distributed area is fallen into according to it, determine the source of heroin.

Based on a heroin source place method of discrimination that is neutral in heroin or acid impurities, be applicable to the heroin ownership differentiating " Southeast Asia " and " South-West Asia " two source places, it is characterized in that:

Step 1) build the partial least squares discriminant analysis function (PLS-DA) in heroin source place, the expression formula of this function is formula (1):

$t=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{\mathrm{il}}{\tilde{x}}_i$ Formula (1)

Wherein $x_i=\frac{A_i}{\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{A}_j}$ Formula (2)

${\tilde{x}}_i=\frac{x_i-\stackrel{\‾}{x_i}}{{\mathrm{\σ}}_i}$ Formula (3)

Wherein a _i1represent 25 neutrality or correlation coefficient value corresponding to acid impurities variable, and σ _irepresent mean value and the standard deviation of 75 parts " Southeast Asia " and each argument value of 75 parts of " South-West Asia " heroin maternal sample after peak area normalization respectively, its numerical value is specifically in table 3; A _irepresent 25 neutrality or peak area numerical value corresponding to acid impurities compound, x _irepresent each variate-value after peak area normalization, m=25;

Step 2), the t distributed area of 150 parts of heroin maternal sample is defined as the classification thresholds of discrimination model, each 75 parts of " Southeast Asia " and " South-West Asia " heroin maternal sample in above-mentioned 150 parts of heroin maternal sample, see Fig. 1: " Southeast Asia " and " South-West Asia " heroin PLS ?DA classification chart, met as can be seen from the t of " Southeast Asia " heroin in Fig. 1: 0<t<6, the t of " South-West Asia " heroin meet: ?6<t<0;

Step 3), calculate the unknown t numerical value capturing heroin sample;

Step 4), determine its distributed areas on " Southeast Asia " and " South-West Asia " heroin PLS-DA classification chart, realize the discriminant classification in its source place accordingly.

In an embodiment of the present invention, the PLS-DA in described heroin source place sentences method for distinguishing and comprises:

Described step 1) in the method for partial least squares discriminant analysis function (PLS-DA) in heroin source place comprise:

First, detected the peak area numerical value obtaining 25 kinds of neutrality or acid impurities compound in heroin sample by Ultra Performance Liquid Chromatography-series connection quadrupole rod time-of-flight mass spectrometry (TOFMS) (UPLC-Q-TOF), comprise the following steps:

(1) extraction of neutrality to be measured or acid impurities in heroin sample

Take heroin sample, be placed in glass centrifuge tube, add acidified solvent, ultrasonic process, adds Extraction solvent, and concussion is centrifugal after extracting, and through filter membrane process, obtains sample solution.

Reagent blank is done in company with sample.

(2) mensuration of neutrality or acid impurities

Set corresponding liquid phase chromatogram condition and Mass Spectrometry Conditions, comprise the molecular formula of setting, retention time, secondary fragment ion parameters, by Ultra Performance Liquid Chromatography-series connection level Four bar time-of-flight mass spectrometry (TOFMS) (UPLC-Q-TOF), 25 kinds of neutrality or the acid impurities compound for the treatment of side carry out Integral Processing, obtain the peak area numerical value of each compound.

Preferably, acidified solvent is the H of 0.25M ₂sO ₄solution, Extraction solvent is sherwood oil: the mixed liquor of methylene chloride=3mL:2mL, and ultrasonic time is 10min, and concussion extraction time is 30min, and centrifugal speed is 5000r/min.

Preferably, centrifugation time is 5-20min, more preferably 10min.

Preferably, filter membrane is 0.22 μm of organic system filter membrane.

Liquid phase chromatogram condition is:

Chromatographic column: Agilent Eclipse Plus C ₁₈rRHD (1.8 μm, 2.1 × 100mm); Sample introduction concentration: 40mg/mL; Sampling volume: 0.5 μ L; Column temperature: 40 DEG C; Mobile phase A: acetonitrile, Mobile phase B: 10mmol/L ammonium formate solution; Flow velocity: 0.4mL/min.Gradient elution program is in table 1.

Table 1 gradient elution program

Time/min	A(％)	B(％)
			0	30	70
1	30	70
			10	55	45
11	95	5
			12	95	5
12.5	30	70
			17	30	70

Mass Spectrometry Conditions is:

First mass spectrometric.Adopt ESI ion gun, positive ion mode; Ion spray voltage: 5.5kV; Remove a bunch voltage (DP): 80V; Vapo(u)rizing temperature: 600 DEG C; Atomization gas (N ₂, GS ₁) the front pressure of post: 50psi; Desolventizing gas (N ₂, GS ₂) the front pressure of post: 50psi; Gas curtain gas (N ₂, CUR) and the front pressure of post: 30psi; Collision energy (CE): 10V; Mass axes: 100 ~ 1000amu.

Second order ms.Adopt ESI ion gun, positive ion mode; MS/MS switched scan (IDA); Ion spray voltage: 5.5kV; Remove a bunch voltage (DP): 80V; Vapo(u)rizing temperature: 600 DEG C; Atomization gas (N ₂, GS ₁) the front pressure of post: 50psi; Desolventizing gas (N ₂, GS ₂) the front pressure of post: 50psi; Gas curtain gas (N ₂, CUR) and the front pressure of post: 30psi; Collision energy (CE): 40V, collision energy expansion (CES): 15V; Mass axes: 100 ~ 1000amu.

Under corresponding Parameter Conditions, see table 2, Integral Processing is carried out to 25 kinds of neutrality or acid impurities compound, obtain the peak area numerical value of each compound.

The liquid matter parameter list of table 2 25 kinds of neutrality or each compound of acid impurities

In table 2, neutrality or its title of acid impurities variable of known structure represent, the neutrality of unknown structure or acid impurities variable represent with " extracting mass number/retention time " ".

Secondly, according to formula (2) and formula (3), the peak area data to each neutrality or acid impurities compound carry out data prediction;

Again, substitute into formula (1) by through the neutrality of data prediction or acid impurities compound data, calculate t numerical value.

In an embodiment of the present invention, the checking of PLS-DA discriminant function is also comprised:

First, the back substitution carrying out training sample differentiates, obtains back substitution and just sentences rate;

Secondly, the differentiation effect of external certificate sample checking discriminant function is adopted.

Partial least square method techniques of discriminant analysis (PLS-DA) is a kind of mode identification method having supervision.Have the mode identification method of supervision to refer to the variable of one group of known class as training set, and obtain discrimination model by this training set, this discrimination model can realize the differentiation to " unknown sample " classification.PLS-DA method is a kind of discriminant analysis method returned based on PLS, considers class members's information that companion matrix provides with code form when structural factor.Therefore, PLS-DA method shows more efficient distinguishing ability with other mode identification method compared with " the soft mode method of independence (SIMCA) of bunch class ".PLS-DA method be also advantageous in that the model quality that it produces is good, and software easily obtains, make it realize fairly simple.

In heroin field, the present invention proposes the heroin source place method of discrimination based on neutral or acid impurities and PLS-DA first, and the technology proposed obviously is better than having method at present.Existing based on method of discrimination that is neutral or acid impurities, be all based on the neutral or presence or absence of acid impurities and the height of content, all do not adopt any chemometrics method.(1) advantage part of the present invention is: the discriminant classification first PLD-DA techniques of discriminant analysis being applied to the heroin source place based on neutral or acid impurities.(2) the present invention PLS-DA techniques of discriminant analysis one step location, fast, effectively of applying.Namely the present invention adopts the PLS-DA module under SIMCA-P statistical software to complete, and carries out without the need to point many analysis modules.(3) this discrimination technology is ripe, and analysis result is objective, and method is simple, easy to utilize.

The rate of just sentencing of the discriminant function that the present invention proposes reaches 99%, differentiates respond well.And it is simple, only need know the peak area data of 25 kinds of neutrality or acid impurities in heroin sample to be sentenced, through simple computation can realize heroin " Southeast Asia ", " South-West Asia " two source place quick, effectively distinguish, the requirement that heroin source place in enormous quantities Quick is analyzed can be met.

Accompanying drawing illustrates:

Fig. 1 " Southeast Asia " and " South-West Asia " heroin PLS-DA classification chart.

Embodiment

The present embodiment provides a kind of based on neutral or acid impurities and PLS-DA heroin source place method of discrimination, be applicable to the heroin ownership differentiating " Southeast Asia " and " South-West Asia " two source places, build the PLS-DA discriminant function in heroin source place, the expression formula of this function is formula (1):

$t=\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}{a}_{\mathrm{il}}{\tilde{x}}_i$ Formula (1)

Wherein $x_i=\frac{A_i}{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}{A}_j}$ Formula (2)

${\tilde{x}}_i=\frac{x_i-\stackrel{\&OverBar;}{x_i}}{{\mathrm{\&sigma;}}_i}$ Formula (3)

Wherein a _i1show 25 neutrality or correlation coefficient value corresponding to acid impurities variable, and σ _irepresent mean value and the standard deviation of 75 parts " Southeast Asia " and each argument value of 75 parts of " South-West Asia " heroin maternal sample after peak area normalization respectively, its numerical value is specifically in table 3; A _irepresent each neutrality or peak area numerical value corresponding to acid impurities compound, x _irepresent each variate-value after peak area normalization.

The t distributed area of 150 parts of heroin maternal sample (each 75 parts of " Southeast Asia " and " South-West Asia " heroin maternal sample) is defined as the classification thresholds of discrimination model, namely as 0<t<6, judge that heroin source place is as " Southeast Asia ", as-6<t<0, judge that heroin source place is as " South-West Asia ".Calculate the unknown t numerical value capturing heroin sample, realize the discriminant classification in its source place accordingly.

Each mean variable value of the neutral or acid impurities PLS-DA discriminant classification model of table 3, standard deviation and factor correlativity coefficient value

In upper table, neutrality or its title of acid impurities variable of known structure represent, the neutrality of unknown structure or acid impurities variable represent with " extracting mass number/retention time " ".

Embodiment 1

For the heroin sample S1 in known source place, first Ultra Performance Liquid Chromatography-series connection quadrupole rod flight time mass spectrum (UPLC-Q-TOF) method is adopted to detect the carrying out of 25 kinds of neutrality or acid impurities in sample S1, obtain the numerical value of each neutrality or acid impurities peak area, specifically see table 4, wherein A ₁-A ₂₅corresponding to the impurity of numbering 1-25 in table 2.

In Ultra Performance Liquid Chromatography-series connection quadrupole rod time-of-flight mass spectrometry (TOFMS) (UPLC-Q-TOF) Simultaneously test heroin, the method for 25 kinds of neutrality or acid impurities comprises the following steps:

(1) extraction of neutrality to be measured or acid impurities in heroin sample

Take " Southeast Asia " heroin sample 100mg, be placed in 8mL glass centrifuge tube, pipette 2mL acidified solvent in glass centrifuge tube with 5mL bottleneck pipettor, acidified solvent is: 0.25M H ₂sO ₄solution, ultrasonic 10min.Pipette 2.5mL Extraction solvent with 5mL bottleneck pipettor and add above-mentioned glass centrifuge tube, Extraction solvent is: sherwood oil: the mixed liquor of methylene chloride=3mL:2mL, and 30min is extracted in concussion.With the centrifugal 10min of 5000r/min rotating speed.Pipette upper organic phase 1mL, avoid carrying water layer secretly, cross 0.22 μm of organic system filter membrane, be transferred in 1.5mL sample injection bottle, sample introduction analysis.

Reagent blank is done in company with sample.

(2) mensuration of neutrality or acid impurities

Sample solution sample introduction analysis under the liquid phase and Mass Spectrometry Conditions of the Ultra Performance Liquid Chromatography-series connection quadrupole rod time-of-flight mass spectrometry (TOFMS) (UPLC-Q-TOF) of setting.

The liquid chromatography parameter set in method is as follows:

Chromatographic column: Agilent Eclipse Plus C ₁₈rRHD (1.8 μm, 2.1 × 100mm); Sample introduction concentration: 40mg/mL; Sampling volume: 0.5 μ L; Column temperature: 40 DEG C; Mobile phase A: acetonitrile, Mobile phase B: ammonium formate (10mmol/L); Flow velocity: 0.4mL/min.Gradient elution program is in table 1.

The Mass Spectrometry Conditions parameter set in method is as follows:

First mass spectrometric.Adopt ESI ion gun, positive ion mode; Ion spray voltage: 5.5kV; Remove a bunch voltage (DP): 80V; Vapo(u)rizing temperature: 600 DEG C; Atomization gas (N ₂, GS ₁) the front pressure of post: 50psi; Desolventizing gas (N ₂, GS ₂) the front pressure of post: 50psi; Gas curtain gas (N ₂, CUR) and the front pressure of post: 30psi; Collision energy (CE): 10V; Mass axes: 100 ~ 1000amu.

Second order ms.Adopt ESI ion gun, positive ion mode; MS/MS switched scan (IDA); Ion spray voltage: 5.5kV; Remove a bunch voltage (DP): 80V; Vapo(u)rizing temperature: 600 DEG C; Atomization gas (N ₂, GS ₁) the front pressure of post: 50psi; Desolventizing gas (N ₂, GS ₂) the front pressure of post: 50psi; Gas curtain gas (N ₂, CUR) and the front pressure of post: 30psi; Collision energy (CE): 40V, (CES): 15V; Mass axes: 100 ~ 1000amu.

Under the molecular formula set, retention time, secondary fragment ion parameters condition, see table 2, Integral Processing is carried out to given 25 kinds of neutrality or acid impurities compound, obtains the peak area numerical value of each compound.

The numerical value of S1 sample 25 neutrality or acid impurities is substituted into discriminant function t, obtains t (S1)=0.737; Due to 0<t<6, therefore this source place of capturing heroin sample is identified as in " Southeast Asia ".

Embodiment 2

For the heroin sample S2 in known source place, first Ultra Performance Liquid Chromatography-series connection quadrupole rod flight time mass spectrum (UPLC-Q-TOF) method is adopted to detect the carrying out of 25 kinds of neutrality or acid impurities in sample S2, obtain the numerical value of each neutrality or acid impurities peak area, specifically see table 4.

From the different of embodiment 1, embodiment 2 is only that known sample derives from South-West Asia.

The numerical value of 25 of S2 sample neutrality or acid impurities is substituted into discriminant function t, obtains t (S2)=-5.116; Due to-6<t<0, therefore this source place of capturing heroin sample is identified as " South-West Asia ".

25 kinds of neutrality of table 4 two parts of heroin samples (S1 and S2) or acid impurities peak area number list

	A 1	A 2	A 3	A 4	A 5	A 6	A 7	A 8	A 9
										S1	14879	21067	2281	195	4547	424	44380	2834	70267
S2	1680	231212	30158	35784	176	7365	8006	62954	221072
											A 10	A 11	A 12	A 13	A 14	A 15	A 16	A 17	A 18
S1	3405	175497	517	44516	8508	1718	8759	5452	2421
										S2	60550	563090	12925	5853	3090	4667	1616	137	890
	A 19	A 20	A 21	A 22	A 23	A 24	A 25

S1	1786	1039	1846	1412	258	46	203
										S2	2383	63923	99268	196622	6513	3199	7089

Utilize the heroin sample of the method to 100 parts of known original producton locations to carry out analysis and distinguishing, only No. 30 sample result of determination mistake, result of determination accuracy rate is 99%, and concrete data are in table 5.

The heroin sample analysis result in table 5 100 parts of known original producton locations

Here be noted that, key of the present invention is to propose a kind of PLS-DA discriminant function, this function is substituted into by described independent variable, differentiation result can be drawn, here discrimination standard is met in order to allow those skilled in the art better understand above-mentioned PLS-DA discriminant function, below the building mode of this function and verification mode are simply introduced, be noted that this structure principle is not intended to limit the invention the technical scheme of method of discrimination.

The method for building up of the PLS-DA discriminant function in described heroin source place comprises:

First, adopt Ultra Performance Liquid Chromatography-series connection quadrupole rod flight time mass spectrum (UPLC-Q-TOF) method to detect the carrying out of 25 kinds of neutrality or acid impurities in 75 parts " Southeast Asia " and 75 parts of " South-West Asia " heroin maternal sample, obtain the numerical value of each neutrality or acid impurities peak area.

Secondly, the peak area numerical value of 25 kinds of neutrality in 150 maternal heroin or acid impurities is substituted into the pre-service that formula (2) and formula (3) carry out data, then the PLS module of SIMCA-P software is adopted to set up PLS-DA discriminant classification model, and obtain the factor correlativity coefficient value of each variable, obtain PLS-DA discriminant function formula (1) accordingly.

The checking of this discriminant function comprises:

First, carry out training the back substitution of sample to differentiate, obtain back substitution and just sentence rate; By all training samples as new samples, in the discriminant function that back substitution has been set up one by one, differentiate ownership.

Secondly, adopt the differentiation effect of external certificate sample checking discriminant function, namely verify that whether discriminant function classification is correct, whether discriminant function has actual application value.Generally just sentencing rate more than 99%, discriminant function has using value.

The foregoing is only preferred embodiment of the present invention, all equalizations done according to the present patent application the scope of the claims change and modify, and all should belong to covering scope of the present invention.

Claims (4)

1. the heroin source place method of discrimination based on neutrality or acid impurities in heroin, measured by centering or acid impurities, determine the source of heroin, it is characterized in that: detected the peak area numerical value obtaining 25 kinds of neutrality or acid impurities compound in heroin sample by Ultra Performance Liquid Chromatography-series connection quadrupole rod time-of-flight mass spectrometry (TOFMS) (UPLC-Q-TOF), process is carried out to above-mentioned peak area numerical value and obtains threshold values of classifying, determine Southeast Asia, heroin source place, the distributed area of South-West Asia, calculate the numerical value of unknown heroin sample, Southeast Asia or South-West Asia distributed area is fallen into according to it, determine the source of heroin.

2., based on a heroin source place method of discrimination that is neutral in heroin or acid impurities, be applicable to the heroin ownership differentiating " Southeast Asia " and " South-West Asia " two source places, it is characterized in that:

Step 1) build the partial least squares discriminant analysis function (PLS-DA) in heroin source place, the expression formula of this function is formula (1):

$t=\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}{a}_{\mathrm{il}}{\tilde{x}}_i$ Formula (1)

Wherein $x_i=\frac{A_i}{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}{A}_j}$ Formula (2)

${\tilde{x}}_i=\frac{x_i-\stackrel{-}{x_i}}{{\mathrm{\&sigma;}}_i}$ Formula (3)

Wherein a _i1show 25 neutrality or correlation coefficient value corresponding to acid impurities variable, i is the natural number of 1-25, and σ _irepresent mean value and the standard deviation of 75 parts " Southeast Asia " and each argument value of 75 parts of " South-West Asia " heroin maternal sample after peak area normalization respectively, its numerical value specifically sees the following form 3; A _irepresent 25 neutrality or peak area numerical value corresponding to acid impurities compound, x _irepresent each variate-value after peak area normalization, m=25;

Step 2), the t distributed area of 150 parts of heroin maternal sample is defined as the classification thresholds of discrimination model, each 75 parts of " Southeast Asia " and " South-West Asia " heroin maternal sample in above-mentioned 150 parts of heroin maternal sample, can find out that the t of " Southeast Asia " heroin meets: 0<t<6, the t of " South-West Asia " heroin meet: ?6<t<0;

Step 3), calculate the unknown t numerical value capturing heroin sample;

Step 4), determine its distributed areas on " Southeast Asia " and " South-West Asia " heroin PLS-DA classification chart, realize the discriminant classification in its source place accordingly,

Each mean variable value of the neutral or acid impurities PLS-DA discriminant classification model of table 3, standard deviation and factor correlativity coefficient value

In upper table, the neutrality of known structure or its title of acid impurities variable represent, the neutrality of unknown structure or acid impurities variable are with " extracting mass number/retention time to represent ".

3. according to claim 2 a kind of based on heroin source place method of discrimination that is neutral or acid impurities, it is characterized in that:

Described step 1) in the method for partial least squares discriminant analysis function (PLS-DA) in heroin source place comprise:

First, detected the peak area numerical value obtaining 25 kinds of neutrality or acid impurities compound in heroin sample by Ultra Performance Liquid Chromatography-series connection quadrupole rod time-of-flight mass spectrometry (TOFMS) (UPLC-Q-TOF), comprise the following steps:

(1) extraction of neutrality to be measured or acid impurities in heroin sample

Take heroin sample, be placed in glass centrifuge tube, add acidified solvent, ultrasonic process, adds Extraction solvent, and concussion is centrifugal after extracting, and through filter membrane process, obtains sample solution;

Reagent blank is done in company with sample;

(2) mensuration of neutrality or acid impurities

Set corresponding liquid phase chromatogram condition and Mass Spectrometry Conditions, comprise the molecular formula of setting, retention time, secondary fragment ion parameters, by Ultra Performance Liquid Chromatography-series connection level Four bar time-of-flight mass spectrometry (TOFMS) (UPLC-Q-TOF), 25 kinds of neutrality or the acid impurities compound for the treatment of side carry out Integral Processing, obtain the peak area numerical value of each compound;

Secondly, according to formula (2) and formula (3), the peak area data to each neutrality or acid impurities compound carry out data prediction;

Again, substitute into formula (1) by through the neutrality of data prediction or acid impurities compound data, calculate t numerical value.

4. a kind of based on neutral or acid impurities and PLS-DA heroin source place method of discrimination according to any one of claim 2-3, is characterized in that:

Also comprise the checking of PLS-DA discriminant function:

First, the back substitution carrying out training sample differentiates, obtains back substitution and just sentences rate;

Secondly, the differentiation effect of external certificate sample checking discriminant function is adopted.