Language
实验动物,肿瘤
z-bo
2020-07-16
1055
Experimental animal grouping is a very important part of the experimental design. When the American Zoological Commission requests submission, it is necessary to explain the basic animal information including animal source, strain, living environment, grouping method, etc. However, most of the published articles do not describe the random grouping method of experimental animals, and the editors requested the original data to prove the reliability of the data. It can be seen that grouping is very important. So what are the principles of grouping experimental animals, and how to implement them?
Animal grouping generally needs to follow the principles of control and randomness. There are mainly two questions here. One is how to design the control group? The second is how to ensure that animals are randomly assigned to these groups?
Question 1: How to set up a control group?
In order to ensure the availability of scientific and effective data, the first design principle of grouping is to establish a control group. In general, to observe the influence of a drug or some external factors on experimental animals, three groups of controls need to be set, a blank control group (negative control group, ), a model control group (positive control group), and a positive control group.
Blank control group: refers to the experimental animals without any treatment, the purpose is to highlight the difference between normal animals and diseased animals; if the treatment effect of a certain drug on diabetic mice is studied, this group is not suffering from diabetes Wild-type mice can be given saline or other drug solvents if necessary.
Model control group: that is, an animal model of disease, given a negative treatment; the purpose is to observe the effect of the drug compared with the treatment group and the animal's performance in the disease state; if you want to observe whether a certain drug has a therapeutic effect on breast cancer, the treatment group Animals require intraperitoneal administration, while the model group is injected with the same dose of solvent excluding drugs, such as normal saline;
Positive control group: usually given to drugs that are known to be effective in disease model animals or other effective factors; the purpose is to compare the effectiveness of a test drug with a positive drug; for example: known doxorubicin on breast cancer tumor growth It has an inhibitory effect, so we can use doxorubicin as a positive control to judge the therapeutic effect of this test drug.
So with the control group, how many groups are needed for the experimental design?
Experimental design, in addition to the above control group, it also needs to add a group of processing factors, generally a dose gradient will be set, and it is best to design more than 3 to form a dose effect. In summary, the recommended groupings are: ① blank group, ② model group, ③ model + positive drug model, ④ model + + low-dose drug, ⑤ model + medium-dose drug, ⑥ model + high-dose drug.
Note: If the subject does not need a disease state, then the model group may not be established;
Note: Not every experiment can find a positive control drug, therefore, not every experiment needs a positive control group;
Note: For specific dose design, you can refer to the link at the end of the article (design of drug concentration)
Question 2: How many are suitable for each group?
Animal experiment design should follow the implementation of the "3R principle", including the principle of substitution, reduction and optimization of experimental animals, where reduction means minimizing the number of experimental animals. Checking the literature, there is no absolute requirement for the number of experimental animals, but while reducing, it must meet the statistical requirements. Statistically, it is generally meaningful to have at least 6 available data in each group.
General mice are generally no less than 10;
No less than 6 rats per group;
The higher the level of large animals, the more expensive the price, which can be appropriately reduced according to the situation, but generally not less than 4-5.
Question 3: How to randomly assign?
Random allocation can ensure the scientificity of the data and can ensure that the data is more convincing. Therefore, the grouping of experimental animals should be strictly in accordance with the principle of randomization, so that each animal has the same opportunity to be allocated to each experimental group.
Basically different experimental purposes, experimental objects, and common methods for grouping are: completely random design, random block design, etc. [3-4]. Random number table and random number remainder grouping method need to be used to realize random grouping.
What is a random number table? Also called random number table, a number table composed of ten numbers randomly generated by the computer from 0 to 9, and the probability of the number of occurrences of each number in the table is basically the same. When grouping, you need to borrow this tool to achieve random allocation.
Method one: completely random design
Randomly group animals indiscriminately through a random number table or computer-generated random numbers [5];
Scope of application: All animals must be "homogeneous" or nearly "homogeneous". In other words, all animals' sex, weight and other related factors (such as tumor-bearing volume) are the same, or approximate.
Steps:
Numbering: Number the experimental animals from 1 to N, generally according to body weight;
Obtain random numbers: Starting from any number in the random number table, sequentially obtain N random numbers in the same direction;
Find the remainder: the random number is divided by the number of groups to find the remainder.
Grouping: Grouping by remainder;
Adjustment: If it is not evenly divided, continue to copy a random number in sequence, the divisor becomes the number a in the group with the largest number, and the remainder is taken as the serial number to be drawn (if the division is even, the remainder is a) until it is adjusted to equal .
For example, divide 15 mice into 3 groups: first number the animals according to body weight, then use a random table to continuously copy 15 numbers, and then use the group number 3 to find the remainder, that is, divide by 3, the remainder 1, 2, 3 respectively Delegates enter groups A, B and C. However, there is one more group A (6), check the random number table, the next random number is 55. 55/6 more than 1, so the first case was adjusted from group A to group C.
Advantages: simple operation, and the principle of random distribution can be achieved;
Disadvantages: First, the process is cumbersome, usually the number of each group cannot be equalized in one operation, and adjustment is required; too many numbers are not easy to quickly group; and if there is a large difference between animals, such as a large body weight interval, this is not applicable method.
Method 2: Random block design
However, if a group of mice, 18g small and 30g large, the weight difference is relatively large, if assigned to each group according to the above completely random method, there may be some components to particularly heavy weight, some Components to extremely small. It is difficult to ensure even distribution, so the random block design came into being.
Random block design: called balanced randomization or restricted randomization, that is, a certain number and proportion of animals are divided into several blocks (generally >3), such as animals of similar weight and animals of the same gender, etc. . The length of the block is preferably more than twice the number of groups. Too small is easy to cause randomness. If there are 4 groups, the length of the block should be more than 8.
Example: There are 40 SD rats, weighing 180-220 grams, which need to be divided into 4 groups, that is, 10 rats in each group (A, B, C, B). The animals should be divided into several blocks according to their weight, and then each block Group animals were randomly assigned to each group.
First, weigh the rats and sort them in order;
Division group: try to divide these rats into 4 groups according to body weight (assuming 180~190g/8, 190-200g/12, 200-210g/16, 210-220g/4), that is 4 Nests, the number of each litter is at least greater than 4 in zoning;
Numbering: Number the four litters in sequence, the first litter 1, 2, 3, 4...8, and so on, numbered to 40;
Randomly assigned block group 1: randomly assign these 8 to 4 groups, 2 for each group, the grouping method adopts the above random number table, and allocates to each group according to the remainder of the random number;
Randomly allocate the remaining 3 blocks in the same way as above.
Advantages: Interval grouping will ensure that the biological differences between the experimental groups are minimized, and the principles of consistency and randomness are followed to make the experimental results more statistically significant.
Note: Sometimes we will encounter multiple factors, we need to divide them in order, for example: 40 mice with large differences in weight, male and female are different, at this time we can not only assign according to the weight block, we need to separate the male and female first, Then each layered by weight. That is to say, it is required to be randomly divided into equal groups according to a certain factor, and then divided into several zones according to another factor within the group.
Note: When the number is small, we can manually calculate the grouping; but when the number is too large, we need to use software such as excel or SPSS to save the trouble of consulting the random number table, and a random allocation process can be completed in a few minutes.
In summary, we can see that we need to follow the principles of control and randomization to group experimental animals, so as to ensure that the data is more scientific and statistically significant, it is best not to add subjective factors for grouping. As an important part of experimental design, experimental animal grouping must be carefully considered before the experiment begins. I hope you will realize the importance of random grouping through our article, and you can learn random grouping in practice.