CN112365991B - Method for mining doubt signal facing SRS combined adverse reaction signal - Google Patents
Method for mining doubt signal facing SRS combined adverse reaction signal Download PDFInfo
- Publication number
- CN112365991B CN112365991B CN202011308163.8A CN202011308163A CN112365991B CN 112365991 B CN112365991 B CN 112365991B CN 202011308163 A CN202011308163 A CN 202011308163A CN 112365991 B CN112365991 B CN 112365991B
- Authority
- CN
- China
- Prior art keywords
- signal
- adverse reaction
- srs
- signals
- doubt
- 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.)
- Active
Links
Images
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
- 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
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- 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
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/40—ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage
Abstract
The invention discloses an in-doubt signal mining method for adverse reaction signals of SRS combined medication, which comprises the following steps: obtaining data through SRS to obtain a combined drug adverse reaction signal set X ═ D1,D2,…,Dx}，DxFor determining the drug combination of adverse reaction signals, identifying the in-doubt signals by a signal strength and threshold value judging method, firstly finding out strong in-doubt signals and all corresponding minimum associated signal groups only containing one signal, judging the signal strength of the signals excluding the strong in-doubt signal groups, and obtaining a minimum associated signal group of the signals, namely weak in-doubt signals and a corresponding minimum associated signal group thereof by a threshold value judging method through a greedy strategy; the in-doubt signal can be used as a false positive signal clue, and the incidence relation, namely the minimum incidence signal group, can be used as a confounding factor influencing the adverse reaction risk assessment of the signal, so that the quality of adverse reaction signal analysis is improved.
Description
Technical Field
The invention relates to the technical field of signal mining, in particular to an in-doubt signal mining method for an adverse reaction signal of SRS combined medication.
Background
The study on the interaction of all the combined drugs cannot be completed in the pre-marketing stage due to the factors such as the experimental range, the study object, the time and the like, and many potential drug interactions can be discovered only in the process of large-scale and long-time use after the drugs are marketed. The Spontaneous drug adverse event Reporting System (SRS) provides an important data source for drug interaction signal mining after marketing. For example, the FDA Adverse Event Reporting System (FAERS) in the united states reflects the complexity of medication safety in real life, and it is investigated that 60% of Adverse Event reports have more than one drug, 70% have more than one drug, and 84% have at least 3 total drugs or Adverse events. These features present opportunities and challenges for combination drug adverse reaction signal mining.
In order to break through the dilemma of 'mass data and lack of information', the mainstream researches utilize a computer to discover adverse reaction signals of the combined medication in batches in the spontaneous report system data. The signal contains two elements: drug combination D and target adverse reaction AE. A drug combination is a collection of drugs, which may also be referred to as a combination or co-drug. If combination D is a signal for an adverse reaction AE, it means that the targeted adverse reaction AE may occur when the patient is taking all of the drugs in D at the same time. When AE is determined, the signal can be expressed as a drug combination. It should be noted that the signal is only a clue and does not prove the causal relationship between the drug combination and the adverse reaction, which needs more complete medical experiment and mechanism analysis to confirm that the signal that can be confirmed is called a positive signal and the signal that cannot be confirmed is called a false positive signal. The higher the positive signal percentage in the signal, the higher the quality of signal mining. Wherein the signal strength metric is the key to signal mining. The imbalance measurement is the basic idea of signal metrics, i.e., "imbalance" or "dissimilarity" of an event of interest as compared to other events. The method comprises two major categories, namely a frequency method and a Bayesian method, wherein the frequency method comprises a relative hazard ratio (RR), a ratio report ratio (PRR), a ratio report ratio (ROR) and the like, and the Bayesian method comprises a Bayesian confidence degree progressive neural network (BCPNN), an empirical Bayesian gamma Poisson distribution reduction Method (MGPS), a BCPNN high-dimensional expansion version, an omega contraction measurement method and the like. The methods have advantages, but overall, the common problems of large signal clue quantity and low accuracy (4%) exist, the advantages of large data cannot be truly played, and the defect is more prominent when high-order combined drug adverse reaction signals are mined. The mode of drug combination in real life is very complex, and the known unbalanced measurement method only evaluates the drug combination of the drug combination under the assumption that the drug combination is independent and irrelevant when evaluating the risk of the adverse drug reaction of the signal, and does not consider the influence of other drug combination, so that other drugs (combination) which are daily combined with the positive signal can be evaluated as false positive signals (namely, the drug combination which is actually unrelated to the adverse drug reaction is judged as the adverse drug reaction signal of the combined drug). It is worth noting that it is also possible that the combined influence of several signals leads to the occurrence of false positive signals. The identification of the association relationship between the signals and the association relationship between the signals has great significance for the judgment of false positive signals and the enrichment of signal analysis clues. At present, no similar research work can be used for reference.
Disclosure of Invention
Technical problem to be solved
Aiming at the problems, the invention provides an in-doubt signal mining method facing to an adverse reaction signal of SRS combined medication, the in-doubt signal can be used as a clue of a false positive signal, and the incidence relation, namely the minimum incidence signal group, can be used as a confounding factor influencing the adverse reaction risk assessment of the signal, thereby being beneficial to improving the analysis quality of the adverse reaction signal.
(II) technical scheme
Based on the technical problems, the invention provides an in-doubt signal mining method facing an adverse reaction signal of SRS combined medication, which comprises the following steps:
s1, initializing the association relation set W ═ phi, the strong in-doubt signal set A ═ phi, the weak in-doubt signal set B ═ phi, acquiring data and preprocessing by the drug adverse event spontaneous reporting system SRS, and then screening to obtain the combined drug adverse reaction signal set X ═ D ═ phi1,D2,…,Dx}，DxA drug combination for determining an adverse reaction signal;
s2, traversing X, finding out all the minimum associated signal groups Y containing only one signal in Z- (X- { D } for each signal D in X sequentially through a signal strength and threshold value judging method, putting D into a strong suspected signal set A, and adding the association relation < D, Y > of D and Y into an association relation set W;
s3, traversing U, solving the minimum association signal set Y for each signal D in U, if the solution is successful, putting D into the weak in-doubt signal set B, and adding < D, Y > into the association set W:
s3.1, enabling a signal set U to be X-A, and starting to traverse U;
s3.2, judging whether the traversal U is finished, if so, entering a step S4, and if not, entering a step S3.3;
s3.3, selecting an unprocessed signal D in U, and enabling a signal set ZZ to be U- { D }, and XX to be phi;
s3.4, traversing ZZ, solving all signals with signal intensity higher than D in ZZ, and adding a candidate signal set XX;
s3.5, making t equal to 1, YtPhi, t represents the number of cycle wheels of the minimum associated signal group Y;
s3.6, judging whether XX is an empty set, if not, entering the step S3.7; if yes, entering step S3.2;
s3.7, traversing XX and obtaining the signal D with the maximum adverse reaction report number in the existing signals of XXtAdding a set of signals YtSynthesis of Yt+1Y is determined by a threshold determination methodt+1If the signal is associated with D, Y is determinedt+1The minimum associated signal group of D is associated<D,Yt+1>Adding an association relation set W, putting D into a weak suspicion signal set B, and entering step S3.2; if not, after t is equal to t +1, step S3.6 is executed again;
s4, outputting an association relation set W, a strong suspicion signal set A and a weak suspicion signal set B; each associated signal set Y corresponding to the signals contained in the strong suspicion signal set A only contains one signal, namely one drug combination, and the weak suspicion signal set B contains the weak suspicion signalCorrelation Y corresponding to signalt+1Is the associated signal group containing the least number of signals, and Yt+1The number of signals contained in the set is more than one, namely the number of drug combinations, and the incidence relation in the incidence relation set W<D,Y>Or<D,Yt+1>Represents Y or Yt+1Is a confounding factor of the corresponding signal D, Y or Yt+1The inclusion of a signal, i.e. the drug combination, affects the assessment of the risk of an adverse reaction of signal D, i.e. the drug combination.
Further, step S2 includes the following steps:
s2.1, traversing X, selecting an unprocessed signal D in X, and enabling a signal set Z to be X- { D };
s2.2, traversing Z, sequentially solving all the minimum associated signal groups Y of D only containing one signal from Z through a signal strength and threshold judgment method, adding the association relation < D, Y > of D and Y into an association relation set W, and after traversing Z is finished, adding the signal D which contains the minimum associated signal group Y only containing one signal into a strong suspected signal set A;
and S2.3, judging whether the traversal X is finished, if so, entering the step S3, and otherwise, entering the step S2.1.
Further, step S2.1 further comprises: let tag be 0; step S2.2 comprises the following steps:
s2.2.1, traversing Z, judging whether the traversing Z is finished, if so, entering a step S2.2.5, and if not, entering a step S2.2.2;
s2.2.2, selecting unprocessed signals D in the signal set ZbInquiring the additional information set I, and judging Q (PS, D, AE) CI-<Q(PS,Db,AE).CI-And Q (PS, D, AE) CI+<Q(PS,Db,AE).CI+If not, entering step S2.2.3 if both are true, otherwise entering step S2.2.5;
s2.2.3, let Y ═ DbCalculating CC (D, Y) and QQ (D, Y) based on the SRS data, judging whether CC (D, Y) is less than or equal to theta or QQ (D, Y) is less than or equal to beta, if any inequality is satisfied, entering step S2.2.4, otherwise, entering step S2.2.5;
s2.2.4, let tag be 1, let W be W { < D, Y > }, return to step S2.2.1;
s2.2.5, determining whether tag is equal to 1, if yes, making a ═ u { D }, and going to step S2.3, if no, going to step S2.3;
wherein the additional information set I is a set of Q (PS, D, AE) corresponding to each signal D in X calculated based on SRS data, Q (PS, D, AE) CI-Lower bound CI representing the confidence interval of the signal strength of said Q (PS, D, AE)-，Q(PS,D,AE).CI+An upper bound CI representing the confidence interval of the signal strength of said Q (PS, D, AE)+Q (PS, D, AE) is a signal metric function of D as adverse reaction AE derived from SRS data of the patient set PS; q (D, Y) ═ Q (PS-
Beta is a strong signal strength threshold value for a signal measurement function supporting signal interference rejection;
and in order to support the adverse reaction report statistical function of eliminating signal interference, G (.) represents a medication case function, C (.,) is an adverse reaction report statistical function, and theta is an adverse reaction report number threshold value.
Further, step S3.4 comprises the steps of:
s3.4.1 signal D not processed in the selected signal set ZZaInquiring the additional information set I, and judging Q (PS, D, AE) CI-<Q(PS,Da,AE).CI-And Q (PS, D, AE) CI+<Q(PS,Da,AE).CI+If not, entering step S3.4.2 if both are true, otherwise entering step S3.4.3; wherein the additional information set I is a set of Q (PS, D, AE) corresponding to each signal D in X calculated based on SRS data, Q (PS, D, AE) CI-Lower bound CI representing the confidence interval of the signal strength of said Q (PS, D, AE)-，Q(PS,D,AE).CI+An upper bound CI representing the confidence interval of the signal strength of said Q (PS, D, AE)+Q (PS, D, AE) is D as a signal metric function of adverse reaction AE derived from SRS data of the set of patients PS;
s3.4.2, let XX ═ U [ U ] Da}；
S3.4.3, judging whether traversing ZZ is finished, if yes, entering step S3.5, otherwise, entering step S3.4.1.
Further, step S3.7 comprises the steps of:
s3.7.1 for each signal D in XXcCalculating CC (DU D) based on SRS datac,Yt) From which to solve
S3.7.2, determining CC (DU D)c,Yt) If so, go to step S3.7.3, otherwise, go to step S3.2;
s3.7.3, let Yt+1＝Yt∪{Dt}，XX＝XX-{Dt}；
S3.7.4, calculating CC (D, Y) based on SRS datat+1) And QQ (D, Y)t+1) Judgment of CC (D, Y)t+1) Theta or QQ (D, Y)t+1) Whether the beta is less than or equal to beta is true, if any inequality is true, the step S3.7.5 is executed, otherwise, the step S3.7.6 is executed;
S3.7.5U-shaped unit<D,Yt+1>}, B ═ u { D }, enter step S3.2;
s3.7.6, making t equal to t +1, and going to step S3.6;
wherein the content of the first and second substances,
an adverse reaction report statistical function supporting signal interference elimination is adopted, G (·) represents a medication case function, AE represents an adverse reaction, C (·,.) is an adverse reaction report statistical function, and theta is an adverse reaction report number threshold;
to support a signal metric function that rejects signal interference, Q (,) is the signal metric function and β is the strong signal strength threshold.
Further, the data is obtained from demographic information, medication information and adverse drug reaction information of the spontaneous drug adverse event reporting system SRS.
Further, the method for determining the drug combination with adverse reaction signals comprises the following steps: if the frequency of the target adverse reaction signals AE reported by the medication patients from the medicine combination D is greater than the support degree threshold value; the length of the drug combination D is not more than the drug combination signal length threshold MAX _ D; d obtained from SRS data of PS patient set is used as lower bound Q (PS, D, AE) of signal intensity confidence interval of target adverse reaction AE->A signal strength threshold δ; d is determined to be the drug combination of the targeted adverse reaction signal AE.
Further, the measurement method adopted by the Q (PS, D, AE) comprises RR, PRR, RoR, BCPNN and MGPS.
The invention also discloses a server, comprising:
at least one processor; and at least one memory communicatively coupled to the processor, wherein:
the memorizer stores program instructions which can be executed by the processor, and the processor calls the program instructions to execute the SRS combined adverse drug reaction signal oriented in-doubt signal mining method.
The invention also discloses a non-transitory computer readable storage medium, which stores computer instructions for causing the computer to execute the method for mining the in-doubt signal facing the SRS combined adverse reaction signal.
(III) advantageous effects
The technical scheme of the invention has the following advantages:
(1) the invention provides a signal association relation mining method facing an SRS combined adverse reaction signal, wherein for a given signal D, if a signal set Y with signal intensity higher than that of D exists, after a medication patient of each signal in Y is removed from SRS data, the intensity evaluation of the signal D is lower than expected, the signal D is associated with the signal set Y, and the Y is called an association signal set of the signal D;
(2) in order to reduce the cost of solving the association relationship, the minimum association signal set of the D is mainly solved, the minimum association signal set is respectively solved in two cases through a signal strength judging method and a threshold value judging method, if the minimum association signal set only contains one signal, the D is marked as a strong suspicion signal, and all the minimum association signal sets are solved; for other in-doubt signals, namely weak in-doubt signals, a greedy strategy is adopted to solve a minimum associated signal group; the strategy not only extracts a representative incidence relation, but also avoids exponential calculation overhead;
(3) the invention also judges the empty set and the adverse reaction report statistical function to determine whether to continue or terminate in advance, thereby reducing the calculation amount and improving the operation efficiency.
Drawings
The features and advantages of the present invention will be more clearly understood by reference to the accompanying drawings, which are illustrative and not to be construed as limiting the invention in any way, and in which:
FIG. 1 is a schematic overall flow chart of an in-doubt signal mining method for an adverse reaction signal of SRS combination according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of an in-doubt signal mining method for adverse reaction signals of SRS combination according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating step S2 according to an embodiment of the present invention;
fig. 4 is a flowchart illustrating step S3 according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
The mining method for analyzing the signal incidence relation of the adverse reaction signals of the combined drug bleeding by using the adverse event report file issued by the FAERS is shown in figure 1 and comprises the following steps:
in the stage I, data are acquired and preprocessed through an SRS (sounding reference signal) of a drug adverse event spontaneous reporting system:data are acquired through an SRS (sounding reference signal) of a drug adverse event spontaneous presentation system, data are deduplicated, drug names are normalized, and target adverse reactions AE are selected. This is the basis and precondition for all signal mining efforts. After data preprocessing, the signal is analyzed in the last two stages. For the purposes of the following expressions, the following symbols are defined: let the total patient set in the SRS be PS ═ p1,p2,…,pmAll drugs were assigned DS ═ d1,d2,…,dn}. For ease of presentation, a set P of patients is given1And P2Defining a set subtraction P1-P2＝P1-P1∩P2. Let C (,) denote the adverse effect reporting statistical function, given a set of patients P, C (P, AE) denotes the number of reports of adverse effect AEs from the set of patients P. Let G (.) denote the medication case function, given signals D, G (D) as a set of cases taking medication combination D.
Obtaining adverse event report files published by FAERS 2018 three-quarter data to obtain 420,915 case reports, 1,651,966 medication records (including 51083 medicine names) and 1,329,530 adverse event records (including 11,944 adverse event records), integrating the demographic information, medication information and adverse drug reactions related by case IDs, and carrying out standardized processing on the adverse drug reaction names and the medicine names. Bleeding events were selected as target adverse events AE, involving 194 adverse events (all generalizable to bleeding events but stated differently) for a total of 19,067 bleeding event records.
And II, excavating adverse reaction signals of the combined medicine: for selected SRS data and target adverse reaction AE, a combined drug adverse reaction signal mining method is mined to obtain a combined drug adverse reaction signal set X ═ D1,D2,…,DxIn which the signal D isiIs a collection of a plurality of medicines, when the patient takes D at the same timeiFor each drug in (1), the patient is said to take Di. . Wherein the signal evaluation is jointly determined by the following three parameters: (1) a Support threshold value Support; the frequency of target adverse reaction AE reported by the drug patients from the drug combination D is more than that of SupA port; (2) a length threshold MAX _ D; the length of drug combination D (i.e., the number of drugs contained in D) is not greater than MAX _ D, i.e., | D<MAX _ D; (3) a signal metric function Q (,) and a signal strength threshold δ; the function Q is input into a patient set PS, a drug combination D and a target adverse reaction AE, and Q (P, D, AE) is output as a Confidence Interval (CI) of signal intensity of an AE signal of D obtained from SRS data of the patient set PS, wherein CI is [ CI ]-，CI+]If the confidence interval is lower bound CI->And delta, D is judged as a signal of the target adverse reaction AE. The signal measurement function Q adopts a non-equilibrium measurement method, which is the category of traditional research work, the known measurement methods comprise a relative hazard ratio (RR), a ratio report ratio (PRR), a report ratio (ROR), a Bayesian confidence degree progressive neural network (BCPNN), an empirical Bayesian gamma Poisson distribution reduction Method (MGPS) and the like, and the corresponding relation between Q and a typical method is shown in Table 1. When solving Q (PS, D, AE) by the non-equilibrium metric method, C (g, (D), AE) needs to be calculated based on SRS data. In order to improve the calculation efficiency of the subsequent operation, each signal D in the signal set X has additional information Q (PS, D, AE), and the additional information set is denoted as I.
TABLE 1 exemplary Signal metric method and confidence interval selection therefor
Using FAERS 2018 three-quarter data mining to mine the drug combination bleeding event signals, setting a Support threshold value Support to be 20, a length threshold value MAX _ D to be 2, adopting a report ratio RoR by a signal measurement function Q (·), and setting a signal intensity threshold value delta to be 2, and obtaining 419 drug combination bleeding event signals in total.
And III, excavating an in-doubt signal and an association relation thereof from the combined drug adverse reaction signal set. For a given signal D, if there is a set of signals with signal strength higher than D
Evaluation of the intensity of Signal D after removal of the drug-treated patient for each Signal Y from the SRS dataIf the signal is estimated to be lower than expected, D is associated with Y, Y is the associated signal group of D, and D is marked as the suspect signal. The in-doubt signal can be used as a false positive signal clue, and the associated signal group can be used as a confounding factor influencing the adverse reaction risk assessment, so that the accuracy of the adverse reaction signal analysis is improved. The computational overhead of solving the association relationships overall is exponential, and in order to improve computational efficiency, only the smallest associated signal group of D (i.e., the associated signal group containing the fewest signals) is solved. If the minimum associated signal group of D only contains one signal, marking D as a strong suspected signal, and solving all the minimum associated signal groups. If the minimum associated signal group of D contains not less than 2 signals, then the D is marked as a weak in-doubt signal, and only one minimum associated signal group is solved.
The signal set X obtained by the existing method contains a large number of false positive signals, and the application value of the combination drug adverse reaction signal mining is seriously damaged. The analysis of the invention considers that the adverse reaction patient report shared between the high-intensity signal(s) and the low-intensity signal(s) can pull up the signal intensity evaluation value of the low-intensity signal, which is an important reason for false positive signal. In other words, if there is a signal set Y with a signal strength higher than that of signal D for a given signal D, the evaluation of the strength of signal D after the removal of the medication patient for each signal in Y from the SRS data is lower than expected, which means that D is likely to be a false positive signal.
For ease of presentation, the following symbolic representations are defined. For a given signal D and a signal set Y, defining an adverse reaction report statistical function CC (,) for eliminating signal interference on the basis of the adverse reaction report statistical function C (,), and enabling the signal set Y to be subjected to signal interference elimination
Since the adverse reaction AE does not change at this stage, the parameter AE is omitted from the function CC. Similarly, a signal metric function QQ (D, Y) for eliminating signal interference is defined based on the signal metric function Q (,) so that
On the basis, the definition of the association relationship is given.
The association relationship is as follows: given a signal D, a set of signals
If the following two conditions are satisfied simultaneously, then D is associated with Y, Y is the associated signal group of D, and D is marked as the suspect signal: (i) each signal D in YbIs higher than D, i.e. satisfies Q (PS, D, AE) CI-<Q(PS,Db,AE).CI-And Q (PS, D, AE) CI+<Q(PS,Db,AE).CI+(ii) a (ii) When the patients who took each signal in Y were culled from the SRS data, the adverse reaction report number for signal D was lower than expected, i.e., CC (D, Y) was not greater than the adverse reaction report number threshold θ, or the intensity of signal D was lower than expected, i.e., QQ (D, Y)-Not greater than the correlation analysis signal strength threshold beta.
In the above definition, the condition (i) is determined using additional information of the signal D, and the overhead is small. The condition (ii) requires a judgment based on SRS data, and the overhead is large. By using the condition (i), all signals satisfying the signal stronger than the signal higher than D in X- { D } can be found first, and the set of the signals is marked as a candidate signal set XX of D. If Y is the associated signal group of D, then
Suppose there are h signals in XX, which means that Y has 2h-1With this possibility, the calculation cost for solving all the association relations is large. In general, if Y1，Y2Set of associated signals all D when Y1Is less than Y2By reference to the data analysis commonly used in the Okamm razor principle, it can be assumed that Y is1Is superior to Y2. By using the principle, the application does not solve all the association relations, and only solves the association signal group with the least number of signals (called the minimum association signal group). According to the minimum associated signal group size, dividing the suspected signal into a strong suspected signal and a weak suspected signal, and simplifying the solving of the association relation:
the first step, solving the strong suspected signal and all the minimum associated signal sets. If the minimum associated signal group Y of the signals D only contains one signal (denoted as D)b) D is marked as a strong in-doubt signal. Obviously, when there are h signals in the candidate signal set XX of D, D has at most h different minimum associated signal sets, and the computational overhead is linear. In view of this, for the strong suspect signal D, all the sets of minimum associated signals of D are solved;
and secondly, solving the weak suspicion signal and a minimum associated signal set thereof. To reduce the analysis overhead, strong incumbent signals are removed from the signal set. For the remaining signal D, either not an incumbent signal or a weak incumbent signal, this can be determined by solving for a minimum set of associated signals for D. If the solution is successful, D is marked as a weak in-doubt signal. Solving the minimum associated signal group Y of D on the signal set XX is the NP problem, and in order to reduce the overhead, the method initializes Y as an empty set and adopts a greedy strategy to iteratively increase signals in Y. Let the t-th round of candidate minimum associated signal set be YtIf the signal set XX is an empty set, it indicates that D has no associated signal set, and the solution is finished. Otherwise, the medication patient of each signal in Y is removed from SRS data, and the signal with the most adverse reaction reports shared by D is selected from XX
Deleting D from XXtD istAdding YtTo obtain Yt+1. If D and Y aret+1And if the correlation exists, the solution is finished. Otherwise, let t be t +1 and enter the next cycle.
In summary, given SRS data, a signal measurement method Q, a signal set X, an additional information set I (including signal strength information of each signal), an adverse reaction report number threshold θ, and an association strength threshold β, the method outputs an association set W, a strong in-doubt signal set a, and a weak in-doubt signal set B. As shown in fig. 2, the specific implementation flow is as follows:
s1, initializing to make the association relation set W equal to phi, the strong in-doubt signal set A equal to phi, and the weak in-doubt signal set B equal to phi;
s2, traversing X, finding out all the minimum associated signal groups Y containing only one signal in Z- (X- { D } for each signal D in X sequentially through a signal strength and threshold value judging method, putting D into a strong suspected signal set A, and adding the association relation < D, Y > of D and Y into an association relation set W; as shown in fig. 3, the method comprises the following steps:
s2.1, traversing X, selecting an unprocessed signal D in X, enabling a signal set Z to be X- { D }, and enabling a mark tag to be 0;
s2.2, traversing the signal set Z, and solving all the minimum associated signal sets of the D, wherein the minimum associated signal set only comprises one signal;
s2.2.1, traversing Z, judging whether the traversing Z is finished, if so, entering a step S2.2.5, and if not, entering a step S2.2.2;
s2.2.2, selecting unprocessed signals D in the signal set ZbInquiring the additional information set I, and judging Q (PS, D, AE) CI-<Q(PS,Db,AE).CI-And Q (PS, D, AE) CI+<Q(PS,Db,AE).CI+If not, entering step S2.2.3 if both inequalities are true, otherwise entering step S2.2.5;
s2.2.3, let Y ═ DbCalculating CC (D, Y) and QQ (D, Y) based on the SRS data, judging whether CC (D, Y) is less than or equal to theta or QQ (D, Y) is less than or equal to beta, if any one of the two inequalities is true, entering step S2.2.4, otherwise entering step S2.2.5;
s2.2.4, let tag be 1, and add the association < D, Y > of D and Y into the association set W, i.e. let W ═ W { < D, Y > }, return to step S2.2.1;
s2.2.5, determining whether tag is equal to 1, if yes, adding D into strong challenge signal set a, that is, making a ═ u { D }, and going to step S2.3, if no, going to step S2.3;
s2.3, judging whether the traversal X is finished, if so, entering a step S3, otherwise, entering a step S2.1;
the steps S2.2.2 and S2.2.3 are obtained according to the above-mentioned methods (i) and (ii) for determining the correlation, respectively, as long as there is one signal D in Zb(iii) satisfying judgment methods (i) and (ii), and obtaining the tagThe value 1, the one signal D is a strong in-doubt signal, but the minimum associated signal set Y of the strong in-doubt signal containing only one signal is not limited to one.
S3, traversing U, solving a minimum associated signal group Y for each signal D in U, if the solving is successful, putting D into a weak in-doubt signal set B, and adding < D, Y > into an association relation set W; as shown in fig. 4, the method comprises the following steps:
s3.1, enabling a signal set U to be X-A, and starting to traverse U;
s3.2, judging whether the traversal U is finished, if so, entering a step S4, and if not, entering a step S3.3;
s3.3, selecting an unprocessed signal D in U, and enabling a signal set ZZ to be U- { D }, and XX to be phi;
s3.4, traversing the signal set ZZ, solving all signals with the signal intensity higher than D in the ZZ, and adding the signals into the candidate signal set XX;
s3.4.1 signal D not processed in the selected signal set ZZaInquiring the additional information set I, and judging Q (PS, D, AE) CI-<Q(PS,Da,AE).CI-And Q (PS, D, AE) CI+<Q(PS,Da,AE).CI+If not, entering step S3.4.2 if both inequalities are true, otherwise entering step S3.4.3;
s3.4.2, mixing DaBy adding XX, i.e. by letting XX ═ XX ^ U { D-a}；
S3.4.3, judging whether traversing ZZ is finished, if yes, entering step S3.5, otherwise, entering step S3.4.1;
s3.5, making t equal to 1, Yt＝Ф；
S3.6, judging whether XX is an empty set, if not, entering the step S3.7; if yes, entering step S3.2;
if XX is empty, then in step S3.4, no signal with signal intensity higher than D is selected from ZZ, D has no associated signal group, and the selection is terminated.
S3.7, traversing XX and obtaining the signal D with the maximum adverse reaction report number in the existing signals of XXtAdding a set of signals YtSynthesis of Yt+1Passing a threshold valueThe judgment method is to compare Yt+1If the signal is associated with D, Y is determinedt+1The minimum associated signal group of D is associated<D,Yt+1>Adding an association relation set W, putting D into a weak suspicion signal set B, and entering step S3.2; if not, after t is made t +1, re-executing step S3.6;
s3.7.1 for each signal D in XXcCalculating CC (DU D) based on SRS datac,Yt) From which to solve
S3.7.2, determining CC (DU D)c,Yt) If so, go to step S3.7.3, otherwise, indicate that D has no associated signal group, terminate the screening, go to step S3.2;
s3.7.3, mixing DtAdding YtTo obtain Yt+1Instant Yt+1＝Yt∪{DtH, mixing D withtDeleted from XX, i.e. let XX be XX- { Dt}；
S3.7.4, calculating CC (D, Y) based on SRS datat+1) And QQ (D, Y)t+1) Judgment of CC (D, Y)t+1) Theta or QQ (D, Y)t+1) Whether beta is less than or equal to beta or not, if any one of the two inequalities is true, D and Y are indicatedt+1Associating, entering step S3.7.5, otherwise entering step S3.7.6;
s3.7.5, mixing D with Yt+1In a relational database<D,Yt+1>Joining in an association set W, i.e. W ═ W & -<D,Yt+1>Adding D into weak doubt signal set B, that is, B ═ U { D }, entering step S3.2;
s3.7.6, let t be t +1, and the process proceeds to step S3.6.
Screening all signals with the signal intensity higher than that of D in U into XX in the step S3.4, performing first-step judgment on the correlation signals, and adding the signals in XX into the set Y in the sequence from large to small of the adverse reaction report number obtained in the steps S3.5 to S3.7 to perform second-step judgment on the correlation signals; and step S3.6 and step S3.7.2 exclude the case that D has no associated signal group, and end the screening, reducing the amount of calculation.
S4, outputting an association relation set W, a strong suspicion signal set A and a weak suspicion signal set B.
The method comprises the steps of carrying out association relation mining on 419 drug combination bleeding event signals obtained by traditional drug combination adverse reaction signal mining, enabling a signal measurement function Q (·) to adopt a report ratio RoR, an adverse reaction report number threshold theta which is 4 and an association relation strength threshold beta which is 1, obtaining 236 strong in-doubt signals and corresponding association relations 3015 in total, and obtaining 11 weak in-doubt signals and corresponding association relations 22. These results help identify false positive signals, such as signal strength, evaluated using 2018 three quarters of SRS data, Q (PS, { glulisine }, AE) ═ 3.13,4.85]Q (PS, { coumarine }, AE) ═ 4.16,5.82]. However, when cases shared by both signals were removed from the SRS data, { glulisine } was less than expected, i.e., QQ ({ glulisine }, { { hydrocinnamol } })-＝0.48<Beta is used as the reference. Thus, { insulin glulisine } is labeled as a strong in-doubt signal and { { coumarine } } is its smallest associated signal set. This result can be corroborated with current results of adverse drug reactions studies, that coumarine is an anticoagulant and is a confirmed bleeding event signal (positive signal); however, it is not reported that glulisine is not reported in this respect, and discussed by two clinicians, it is recognized that { glulisine } is a false positive signal, and it is pointed out that if the signal { glulisine } is further verified, coumarine should be incorporated into the experimental design as a confounding factor. A bleeding event prompting signal in a drug knowledge base drug bank and scientific research libraries PubMed and Medline is selected as a positive signal base, and the positive signal base is not used in combination adverse reaction signal mining and the invention, so that the positive signal ratio can be used for evaluating the signal quality under the same zero knowledge condition. There were 53 positive signals in total (12.7% positive signal percentage) out of the 419 signals from stage II. The present invention obtains 1 positive signal (0.4%) in 247 suspected signals. Therefore, the method is beneficial to identifying the false positive signals and improving the signal quality.
The method can be used for analyzing the combined adverse reaction signals independently, and can also be used in cooperation with a strong signal screening method facing the SRS combined adverse reaction signals and/or an in-doubt signal mining method facing the SRS combined adverse reaction signals. The three methods are all used for analyzing the relation between signals, but the targets are respectively emphasized, strong signal screening is used for finding out a positive signal clue, doubtful signal mining is used for finding out a false positive signal clue, equivalence relation signal mining is used for finding out a signal clue with consistent data height, and the clues are respectively beneficial to improving the signal analysis quality. And (4) carrying out equivalent relation signal mining firstly, then carrying out strong signal screening and doubt signal mining execution. The equivalence relation signal mining method obtains a maximum equivalence set, only one signal in each maximum equivalence set needs to be selected to participate in subsequent analysis, and the calculation cost of the whole analysis process can be reduced.
Finally, it should be noted that the above-described method can be converted into software program instructions, and can be implemented by using a control system including a processor and a memory, or by using computer instructions stored in a non-transitory computer-readable storage medium. The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute some steps of the methods according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
In summary, the mining method for the signal association relationship facing the adverse reaction signal of the SRS combined medication has the following advantages:
(1) the invention provides a signal association relation mining method facing an SRS combined adverse reaction signal, wherein for a given signal D, if a signal set Y with signal intensity higher than that of D exists, after a medication patient of each signal in Y is removed from SRS data, the intensity evaluation of the signal D is lower than expected, the signal D is associated with the signal set Y, and the Y is called an association signal set of the signal D;
(2) in order to reduce the cost of solving the association relationship, the minimum association signal set of the D is mainly solved, the minimum association signal set is respectively solved in two cases through a signal strength judging method and a threshold value judging method, if the minimum association signal set only contains one signal, the D is marked as a strong suspicion signal, and all the minimum association signal sets are solved; for other in-doubt signals, namely weak in-doubt signals, a greedy strategy is adopted to solve a minimum associated signal group; the strategy not only extracts a representative incidence relation, but also avoids exponential calculation overhead;
(3) the invention also judges the empty set and the adverse reaction report statistical function to determine whether to continue or terminate in advance, thereby reducing the calculation amount and improving the operation efficiency.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.
Claims (10)
1. A method for mining an in-doubt signal facing an adverse reaction signal of SRS combined medication is characterized by comprising the following steps:
s1, initializing the association relation set W ═ phi, the strong in-doubt signal set A ═ phi, the weak in-doubt signal set B ═ phi, acquiring data and preprocessing by the drug adverse event spontaneous reporting system SRS, and then screening to obtain the combined drug adverse reaction signal set X ═ D ═ phi1,D2,…,Dx}，DxA drug combination for determining an adverse reaction signal;
s2, traversing X, finding out all the minimum associated signal groups Y containing only one signal in Z- (X- { D } for each signal D in X sequentially through a signal strength and threshold value judging method, putting D into a strong suspected signal set A, and adding the association relation < D, Y > of D and Y into an association relation set W;
s3, traversing U, solving the minimum association signal set Y for each signal D in U, if the solution is successful, putting D into the weak in-doubt signal set B, and adding < D, Y > into the association set W:
s3.1, enabling a signal set U to be X-A, and starting to traverse U;
s3.2, judging whether the traversal U is finished, if so, entering a step S4, and if not, entering a step S3.3;
s3.3, selecting an unprocessed signal D in U, and enabling a signal set ZZ to be U- { D }, and XX to be phi;
s3.4, traversing ZZ, solving all signals with signal intensity higher than D in ZZ, and adding a candidate signal set XX;
s3.5, making t equal to 1, YtPhi, t represents the number of cycle wheels of the minimum associated signal group Y;
s3.6, judging whether XX is an empty set, if not, entering the step S3.7; if yes, entering step S3.2;
s3.7, traversing XX and obtaining the signal D with the maximum adverse reaction report number in the existing signals of XXtAdding a set of signals YtSynthesis of Yt+1Y is determined by a threshold determination methodt+1If the signal is the associated signal group of D, Y is determined if the signal is the associated signal group of Dt+1The minimum associated signal group of D is associated<D,Yt+1>Adding an association relation set W, putting D into a weak suspicion signal set B, and entering step S3.2; if not, after t is made t +1, re-executing step S3.6;
s4, outputting an association relation set W, a strong suspicion signal set A and a weak suspicion signal set B; each associated signal set Y corresponding to the signal contained in the strong suspicion signal set A only contains one signal, namely one drug combination, and the association relation Y corresponding to the weak suspicion signal contained in the weak suspicion signal set Bt+1Is the associated signal group containing the least number of signals, and Yt+1The number of signals contained in the set W, i.e. the number of drug combinations, is more than oneAssociation relation<D,Y>Or<D,Yt+1>Represents Y or Yt+1Is a confounding factor of the corresponding signal D, Y or Yt+1The inclusion of a signal, i.e. the drug combination, affects the assessment of the risk of an adverse reaction of signal D, i.e. the drug combination.
2. The method for mining the suspected signal facing the SRS combined adverse reaction signal, according to claim 1, wherein the step S2 includes the steps of:
s2.1, traversing X, selecting an unprocessed signal D in X, and enabling a signal set Z to be X- { D };
s2.2, traversing Z, sequentially solving all the minimum associated signal groups Y of D, which only contain one signal, from Z through a signal strength and threshold judgment method, adding the association relation < D, Y > of D and Y into an association relation set W, and after the traversing Z is finished, adding the signal D, which has the minimum associated signal group Y, which only contains one signal, into a strong doubt signal set A;
and S2.3, judging whether the traversal X is finished, if so, entering the step S3, and otherwise, entering the step S2.1.
3. The method for mining the doubt signal facing the adverse reaction signal of the SRS combined medication, as claimed in claim 2, wherein the step S2.1 further comprises: let tag be 0; step S2.2 comprises the following steps:
s2.2.1, traversing Z, judging whether the traversing Z is finished, if so, entering a step S2.2.5, and if not, entering a step S2.2.2;
s2.2.2, selecting unprocessed signals D in the signal set ZbInquiring the additional information set I, and judging Q (PS, D, AE) CI-<Q(PS,Db,AE).CI-And Q (PS, D, AE) CI+<Q(PS,Db,AE).CI+If not, entering step S2.2.3 if both are true, otherwise entering step S2.2.5;
s2.2.3, let Y ═ DbCalculating CC (D, Y) and QQ (D, Y) based on the SRS data, judging whether CC (D, Y) is less than or equal to theta or QQ (D, Y) is less than or equal to beta, if any inequality is satisfied, entering step S2.2.4, otherwise entering step S2.2.5；
S2.2.4, let tag be 1, let W be W { < D, Y > }, return to step S2.2.1;
s2.2.5, determining whether tag is equal to 1, if yes, making a ═ u { D }, and going to step S2.3, if no, going to step S2.3;
wherein the additional information set I is a set of Q (PS, D, AE) corresponding to each signal D in X calculated based on SRS data, Q (PS, D, AE) CI-Lower bound CI representing the confidence interval of the signal strength of said Q (PS, D, AE)-，Q(PS,D,AE).CI+An upper bound CI representing the confidence interval of the signal strength of said Q (PS, D, AE)+Q (PS, D, AE) is a signal metric function of D as adverse reaction AE derived from SRS data of the patient set PS; q (D, Y) ═ Q (PS-
Beta is a strong signal strength threshold value for a signal measurement function supporting signal interference rejection;
and in order to support the adverse reaction report statistical function of eliminating signal interference, G (.) represents a medication case function, C (.,) is an adverse reaction report statistical function, and theta is an adverse reaction report number threshold value.
4. The method for mining the suspected signal for the SRS combined adverse reaction signal according to claim 1, wherein step S3.4 comprises the steps of:
s3.4.1 signal D not processed in the selected signal set ZZaInquiring the additional information set I, and judging Q (PS, D, AE) CI-<Q(PS,Da,AE).CI-And Q (PS, D, AE) CI+<Q(PS,Da,AE).CI+If not, entering step S3.4.2 if both are true, otherwise entering step S3.4.3; wherein the additional information set I is a set of Q (PS, D, AE) corresponding to each signal D in X calculated based on SRS data, Q (PS, D, AE)-Lower bound CI representing the confidence interval of the signal strength of said Q (PS, D, AE)-，Q(PS,D,AE).CI+An upper bound CI representing the confidence interval of the signal strength of said Q (PS, D, AE)+Q (PS, D, AE) is a signal metric function of D as adverse reaction AE derived from SRS data of the patient set PS;
s3.4.2, let XX ═ XX ^ U { D-a}；
S3.4.3, judging whether traversing ZZ is finished, if yes, entering step S3.5, otherwise, entering step S3.4.1.
5. The method for mining an in-doubt signal facing an adverse reaction signal of SRS combination according to claim 1, wherein step S3.7 includes the steps of:
s3.7.1 for each signal D in XXcCalculating CC (DU D) based on SRS datac,Yt) From which to solve
S3.7.2, determining CC (DU D)c,Yt) If so, go to step S3.7.3, otherwise, go to step S3.2;
s3.7.3, let Yt+1＝Yt∪{Dt}，XX＝XX-{Dt}；
S3.7.4, calculating CC (D, Y) based on SRS datat+1) And QQ (D, Y)t+1) Judgment of CC (D, Y)t+1) Theta or QQ (D, Y)t+1) Whether the beta is less than or equal to the beta is true, if any inequality is true, the step S3.7.5 is entered, otherwise, the step S3.7.6 is entered;
S3.7.5U-shaped unit<D,Yt+1>} B ═ u { D }, go to step S3.2;
s3.7.6, making t equal to t +1, and going to step S3.6;
wherein the content of the first and second substances,
in order to support adverse reaction report statistical function of eliminating signal interference, G (.) represents a medication case function, AE represents adverse reaction, C (.,) is an adverse reaction report statistical function, and theta is an adverse reaction reportAn advertised count threshold;
to support a signal metric function that rejects signal interference, Q (,) is the signal metric function and β is the strong signal strength threshold.
6. The SRS combined adverse drug reaction signal-oriented doubt signal mining method as claimed in claim 1, wherein the data is obtained from demographic information, medication information, adverse drug reaction information of a spontaneous drug adverse event presentation system SRS.
7. The method for mining an in-doubt signal facing an adverse reaction signal of SRS combination according to claim 1, wherein the method for determining the drug combination of the adverse reaction signal comprises: if the frequency of target adverse reaction signals AE reported by the medication patients of the medicine combination D is greater than the support threshold; the length of the drug combination D is not more than the drug combination signal length threshold MAX _ D; d obtained from SRS data of PS patient set is used as lower bound Q (PS, D, AE) of signal intensity confidence interval of target adverse reaction AE->A signal strength threshold δ; d is determined to be the drug combination of the targeted adverse reaction signal AE.
8. The method for mining an in-doubt signal facing an adverse reaction signal of SRS combination according to claim 3, wherein the measurement method adopted by Q (PS, D, AE) comprises RR, PRR, RoR, BCPNN, MGPS.
9. A server, comprising:
at least one processor; and at least one memory communicatively coupled to the processor, wherein:
the memory stores program instructions executable by the processor, the program instructions being invoked by the processor to perform the method of in-doubt signal mining for SRS combination adverse reaction oriented signals as claimed in any one of claims 1 to 8.
10. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method for in-presence signal mining for SRS combination adverse reaction signals according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011308163.8A CN112365991B (en) | 2020-11-19 | 2020-11-19 | Method for mining doubt signal facing SRS combined adverse reaction signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011308163.8A CN112365991B (en) | 2020-11-19 | 2020-11-19 | Method for mining doubt signal facing SRS combined adverse reaction signal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112365991A CN112365991A (en) | 2021-02-12 |
| CN112365991B true CN112365991B (en) | 2022-05-03 |
Family
ID=74533010
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011308163.8A Active CN112365991B (en) | 2020-11-19 | 2020-11-19 | Method for mining doubt signal facing SRS combined adverse reaction signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112365991B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2722395A1 (en) * | 2001-10-15 | 2014-04-23 | Bioarray Solutions Ltd | Multiplexed analysis of polymorphic loci by concurrent interrogation and enzyme-mediated detection |
| CN108538396A (en) * | 2018-04-09 | 2018-09-14 | 南京邮电大学 | A kind of adverse drug reaction shadowing effect removing method based on mixed model |
| CN109215799A (en) * | 2018-08-20 | 2019-01-15 | 南京邮电大学 | Screening technique for correlation signal false in drug combination adverse reaction data reporting |
-
2020
- 2020-11-19 CN CN202011308163.8A patent/CN112365991B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2722395A1 (en) * | 2001-10-15 | 2014-04-23 | Bioarray Solutions Ltd | Multiplexed analysis of polymorphic loci by concurrent interrogation and enzyme-mediated detection |
| CN108538396A (en) * | 2018-04-09 | 2018-09-14 | 南京邮电大学 | A kind of adverse drug reaction shadowing effect removing method based on mixed model |
| CN109215799A (en) * | 2018-08-20 | 2019-01-15 | 南京邮电大学 | Screening technique for correlation signal false in drug combination adverse reaction data reporting |
Non-Patent Citations (3)
| Title |
|---|
| 药品不良反应信号检测中的关联规则;王超等;《中国药物警戒》;20100315(第03期);第28-33页 * |
| 药品不良反应信号检测中遮蔽效应的影响及消除方法;叶小飞等;《中国药物警戒》;20130115(第01期);第66-71页 * |
| 阿奇霉素不良反应信号挖掘与评价;冷维萍等;《药物流行病学杂志》;20200315(第03期);第96-105页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112365991A (en) | 2021-02-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11954614B2 (en) | Systems and methods for visualizing a pattern in a dataset | |
| Nowicka et al. | CyTOF workflow: differential discovery in high-throughput high-dimensional cytometry datasets | |
| Ashhurst et al. | Integration, exploration, and analysis of high‐dimensional single‐cell cytometry data using Spectre | |
| JP7057913B2 (en) | Big data analysis method and mass spectrometry system using the analysis method | |
| US20180225416A1 (en) | Systems and methods for visualizing a pattern in a dataset | |
| JP6715451B2 (en) | Mass spectrum analysis system, method and program | |
| Bock et al. | Association mapping in biomedical time series via statistically significant shapelet mining | |
| CN112289376B (en) | Method and device for detecting somatic cell mutation | |
| Spinozzi et al. | VISPA2: a scalable pipeline for high-throughput identification and annotation of vector integration sites | |
| Cordero et al. | Tracing co-regulatory network dynamics in noisy, single-cell transcriptome trajectories | |
| CN112365991B (en) | Method for mining doubt signal facing SRS combined adverse reaction signal | |
| Azad et al. | Immunophenotype discovery, hierarchical organization, and template-based classification of flow cytometry samples | |
| US20010056571A1 (en) | Difference engine method and apparatus | |
| AU2020356582A1 (en) | Single cell RNA-seq data processing | |
| US20200265919A1 (en) | Analyzing device, analysis method program, and non-volatile storage medium | |
| CN112365990B (en) | Strong signal screening method for adverse reaction signals of SRS combined medication | |
| CN112365989B (en) | Equivalent signal mining method for SRS combined adverse reaction signals | |
| Cisar et al. | A unified pipeline for FISH spatial transcriptomics | |
| CN110970089A (en) | Preprocessing method and preprocessing device for fetal concentration calculation and application of preprocessing method and device | |
| Pura | Multiple Testing Embedded in an Aggregation Tree With Applications to Omics Data | |
| Lauria | Rank-based miRNA signatures for early cancer detection | |
| CN115527610B (en) | Cluster analysis method for single-cell histology data | |
| CN116364182A (en) | Integrated analysis method for single cell transcriptome and TCR and BCR sequencing data | |
| CN117893512A (en) | Nucleic acid detection and data analysis method, apparatus, system and storage medium | |
| Keener et al. | A unified pipeline for FISH spatial transcriptomics. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |