| 查看: 740 | 回复: 1 | |||
| 【悬赏金币】回答本帖问题，作者zhuchaoying将赠送您 5 个金币 | |||
[求助]
如何算有效种群Ne 已有1人参与
|
|||
|
为什么我用NeEstimator会出现负值，这是怎么回事 发自小木虫Android客户端 |
回复此楼» 猜你喜欢
导师吐槽：我怎么摊上了这么个极品研究生！
已经有7人回复
-
青基如何破局
已经有7人回复
-
有多少人是今天查系统知道结果的？
已经有17人回复
-
基金不中，共勉
已经有10人回复
-
国自然面上复盘~欢迎讨论
已经有11人回复
-
为什么资助数各大高校都创新高，自己申请怎么就这么难
已经有7人回复
-
基金未中，这种答复是模板吗？
已经有6人回复
-
看板上这么多中的，有点像50人群里49个人都是骗子的那种感觉……
已经有6人回复
-
怎么看青基中了没有啊
已经有5人回复
-
梦想
已经有3人回复
【答案】应助回帖
|
Negative or infinite estimates of Ne Each method estimates Ne based on the magnitude of a genetic characteristic, after accounting for the expected contribution from random errors in sampling. For example, in the temporal and linkage disequilibrium methods, the contribution to or expected to arise from sampling S individuals is approximately 1/S. To obtain an unbiased estimate of the genetic index, this amount attributed to sampling is subtracted from the raw estimate. Because the actual contribution of sampling error is a random variable, it can be smaller than the expected value, and when that happens subtracting the expected contribution can produce a negative estimate of adjusted or , which in turn produces a negative estimate of Ne. This also can occur with unbiased estimators of FST or genetic distance. The usual interpretation in this case is that the estimate of Ne is infinity – that is, there is no evidence for variation in the genetic characteristic caused by genetic drift due to a finite number of parents — it can all be explained by sampling error. In the confidence intervals (CIs), such values are reported as ‘Infinite,’ meaning that the confidence interval includes infinity. However, the point estimates of Ne are reported even if they are negative (in accessory output files), as in some applications this information can be useful. For example, say you have several replicate samples from the same population and use each sample to estimate Ne. An overall estimate of Ne can be obtained by taking the harmonic mean of the separate estimates, even if they are negative. You will get an approximately, but not exactly, correct answer if you replace negative estimates of Ne with infinity before taking a harmonic mean. This issue is discussed in Waples and Do (2010). |
2楼2017-10-20 11:15:06
投票:
10