2009年2月27日星期五

常见的逻辑错误

当你与别人讨论,尝试获得答案或解释时,你可能会遇到一些人犯上逻辑谬误。这样的讨论是无意义的。你可能尝试向对手要求证据或提供其他假设,令你获得更好或更简单的解释。如果都失败,可以尝试指出你讨论对手的问题。你可辨认他的逻辑问题以免深究,以及可告知讨论对手关於他的谬误。以下是简单介绍其中最常见的谬误:

人身攻击(ad hominem):
拉丁语「向着人」的意思。辩者用人身攻击来攻击对手,而不是在讨论议题。当辩者不能用证据、事实或理由去维护他的立场,他可能透过标签、稻草人、骂人、挑衅及愤怒的人身攻击方式来攻击对手。

诉诸无知(appeal to ignorance / argumentum ex silentio):
以诉诸无知作为某些证据。(例如:我们没有证据说神不存在,所以祂一定存在。又例如:由於我们没有关於外星人的知识,这表示他们并不存在。)对某些东西的无知,是与它的存在与否无关。

全知论据(argument from omniscience):
(例如:所有人都相信某些东西,每个人都知道的。)辩者需要有全知能力以清楚每个人的信仰、怀疑或他们的知识。小心如「所有」、「每个人」、「每种东西」、「绝对」等词语。

诉诸信心(appeal to faith):
(例如:如果你不相信,是不能清楚明白的。)如果辩者倚仗信心作为他论据的根基,那麽你在以後的讨论所能得到的将不多。根据定义,「信心」是倚靠相信,并非靠逻辑或证据支持。信心倚赖非理性的思想,并会产生不妥协。

诉诸传统(appeal to tradition):
(类似主流思想谬误)(例如:占星、宗教、奴隶)只因为人们以此为传统,与它本身的存活能力无关。

诉诸权威(argument from authority / argumentum ad verecundiam):
以「专家」或权威的说话作论据的根基,而不是用逻辑或证据来支持该论据。(例如:某某教授相信创造科学。)只由於某个权威的声称,不足以代表他已令这声称正确。假如辩者展示某专家的论据,那麽看看它有否伴随着原因,以及它背後证据的来源。

不良後果论据(argument from adverse consequences):
(例如:我们应判被告有罪,否则其他人会仿效而犯上类似的罪行。)只因为讨厌的罪行或行为出现,并不足以代表被告犯了该罪,或代表我们应判他有罪。(又例如:灾难的出现是因为神惩罚不信者,所以我们都应该信神。)只因灾害或惨剧发生,与神是否存在、或我们该信甚麽并无关系。

恐吓论据(argumentum ad baculum):
论据根基於恐惧或威胁。(例如:如果你不信神,你将会下地狱被火烧。)

无知论据(argumentum ad ignorantiam):
误导性的论据,倚仗於人们的无知。

群众论据(argumentum ad populum):
论据诉诸感性的弱点,而非事实和原因,旨在煽动群众的支持。

主流思想谬误(bandwagon fallacy):
只因为很多人相信或实践,便认为一个思想有价值。(例如:大多数人相信神,所以它一定是真的。)只因为很多人相信某些东西,与那是事实与否并无关系。如很多人在黑死病时期都相信疫症是由於魔鬼引起,有多少人相信跟疫症的起因全无关系。

窃取论点(begging the question):
(例如:我们必须鼓励年青人去崇拜神,以灌输道德行为。)可是宗教与崇拜真的产生道德行为吗?

循环论证(circular reasoning):
陈述某命题,而其实那正是需要被证实的。(例如:神存在是因为圣经有记载,圣经存在是因为神所默示的。)

构成谬误(composition fallacy):
当某论据的结论,是倚靠由某东西从部份至整体、或从整体至部份的错误特性。(例如:人类有意识,而人体和人脑都是由原子组成,所以原子都有意识。又例如:文书处理软件由佷多原位组(byte)组成,所以一个原位组是组成文书处理软件的一部份。)

确认性偏见(confirmation bias):
(类似监视下的选择)这是指一种选择性的思想,集中於支持相信的人已相信的证据,而忽略反驳他们信念的证据。确认性偏见常见於人们以信心、传统及成见为根据的信念。例如,如果有些人相信祈祷的力量,相信的人只会注意到少量「有回应」的祈祷,而忽略大多数无回应的祈祷。(这表示祈祷的价值最差只是随机,最好也只有心理上的安慰作用。)

混淆相关及起因(confusion of correlation and causation):
(例如:玩象棋的人男性比女性多,所以男性棋艺也比女性高。又例如:儿童观看电视的暴力场面,成长後会有暴力倾向。)但是,那是由於电视节目引致暴力,还是有暴力倾向的儿童喜欢观看暴力节目?真正引致暴力的原因可能是完全与电视无关。Stephen
Jay Gould 把相关引致的无效假设称为「可能是人类推理上两三种最严重和最普遍的错误」。

错误二分法/排中(excluded middle / false dichotomy):
只考虑极端。很多人用亚理士多德式(Aristotelian)的「非此即彼」的逻辑去解释上下、黑白、对错、爱恶等。(例如:你若非喜欢它,就是不喜欢它。他如不是有罪,就是无罪。)很多时人们没有看到在两个极端之间出现的连续,这个宇宙也包含很多「可能」的。

隐藏证据(half truths / suppressed evidence):
故意欺骗的陈述,通常隐藏一些事实,而那是构成准确描述所必需的。

暗示/诱导性问题(loaded questions):
问题加入假设,一旦回答便显示了一个暗示性的同意。(例如:你停止了打你的妻子吗?)

无意义的问题(meaningless question):
(例如「上面有多高?」「一切皆可能吗?」)「上面」描述方向,不是可衡量的单位。假如一切都证实可能,那麽「不可能」都可能出现,矛盾便出现。尽管一切不一定证实可能,亦可以有无数的可能和无数的不可能。很多无意思的问题都包含了空废的词语,如"is," "are," "were," "was," "am," "be," 或 "been."

统计性质的误解(misunderstanding the nature of statistics):
(例如:大多数美国人都死在医院内,所以应尽量远离医院。)「统计显示,通常染上进食习惯的人,很少能生存。」-- Wallace Irwin

不当结论(non sequitur):
拉丁语「它没有跟随」的意思。推断或结论没有跟随已建立的前提或证据。(例如:在月圆时出生的人较多。结论:月圆引致出生率上升。)可是,是月圆引致较多出生,还是由於其他原因(可能是统计上的期望差异)?

监视下的选择(observational selection):
(类似确认性偏见)指出有利的,却忽略不利的事实。谁去过拉斯维加斯(Las Vegas)赌场会见到人们在赌桌上和老虎机上赢钱,赌场经理会响钟及鸣笛以公告胜利者,却永不会提及失败者。这可令人觉得胜出的机会看来颇大,但是事实却刚刚相反。

错误因果(post hoc, ergo propter hoc):
拉丁语「它发生在之後,所以它是结果。」与不当结论类似,不过与时间有关。(例如:她去了中国之後病了,所以中国有些东西令到她病。)可能她的病是由於其他原因,与中国完全无关。

证明不存在(proving non-existence):
当辩者无法为他的声称提供证据,他可能会挑战他的对手,叫对手证明他的声称不存在。(例如:证明神不存在;证明不明飞行物体未曾到过地球;等等)尽管有人可以在特定的限制中证明不存在,如在盒中没有某些东西,可是却无法证明普遍性、绝对性或认知性的不存在。无人能证明一些不存在的东西。提出声称的人必需自己证明那声称的存在。

扯开话题(red herring):
辩者改变话题,以分散注意力。

实体化谬误(reification fallacy):
当人们把抽象的信念或假设性的构想,当作是实在的事物。如以IQ题作为真实衡量智慧的方法;由抽象的社会构想而来的种族概念(尽管基因属性的存在),源自经拣选的属性组合,或者标签某一组人;占星;耶稣;圣诞老人;等等。

滑坡谬误(slippery slope):
一个步骤、法律、或行动的改变,可引致不良的後果。(例如:如果我们容许医生帮助安乐死,那麽去到最後,政府会控制我们如何死。)不一定只因为我们的改变,出现了滑坡,便会使预计的後果实现。

片面辩护(special pleading):
以新鲜或特别的声称,抗衡对手的陈述;展示论据时只着重主题中有利或单一的范畴。(例如:神为何在世上创造这麽多苦难?答案是:你必须明白,神自有祂神奇的安排,我们没有特权去知道的。又例如:星座是准确的,但你必须先了解背後的理论。)

小众统计(statistics of small numbers):
类似监视下的选择。(例如:我的父母吸了一世烟,但他们从未患过癌症。又例如:我不管其他人如何讲 Toyota,我的 Toyota 却从未发生过问题。)只指出少量有利数据,与整体机会并无关系。

2009年2月24日星期二

GST pull down实验方案(冷泉港实验室)

GST Pull-down
Margret B. Einarson, Elena N. Pugacheva, and Jason R. Orlinick
This protocol was adapted from "Identification of Protein-Protein Interactions with Glutathione-S-Transferase Fusion Proteins," Chapter 6, in Protein-Protein Interactions, 2nd edition (eds. Golemis and Adams). Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, USA, 2005.


INTRODUCTION

Glutathione-S-transferase (GST) fusion proteins have had a wide range of applications since their introduction as tools for synthesis of recombinant proteins in bacteria. One of these applications is their use as probes for the identification of protein-protein interactions. The pull-down method described in this protocol is fundamentally similar to immunoprecipitation. Immunoprecipitation is based on the ability of an antibody to bind to its antigen in solution, and the subsequent purification of the immunocomplex by collection on protein A- or G-coupled beads. Similarly, the GST pull-down is an affinity capture of one or more proteins (either defined or unknown) in solution by its interaction with the GST fusion probe protein and subsequent isolation of the complex by collection of the interacting proteins through the binding of GST to glutathione-coupled beads.


RELATED INFORMATION

An introduction to the use of GST fusion proteins for studying protein-protein interactions can be found in Identification of Protein-Protein Interactions with Glutathione-S-Transferase (GST) Fusion Proteins.

This protocol is designed to use a 35S-labeled cell lysate as the source for interacting proteins. For 35S-labeling procedures, see Orlinick and Chao (1996) and Spector et al. (1998). Additionally, if the interacting protein of interest is known to be confined to a specific cellular compartment (e.g., the nucleus), a fraction of the cell lysate corresponding to that compartment (e.g., a nuclear extract [to prepare, see Dignam et al. 1982]) can be used in place of a total cell lysate.


MATERIALS

This procedure may require equipment or reagents for Western analysis, Coomassie blue staining, and/or silver staining (see Step 13).

Reagents

Cell lysate (unlabeled or labeled with 35S, depending on experimental goal)

This experiment compares GST versus GST fusion protein, so it is necessary to prepare enough lysate to provide equal amounts of lysate in each reaction. The amount of lysate needed to detect an interaction is highly variable. Start with lysate equivalent to 1 x 106 to 1 x 107 tissue culture cells.

GST fusion protein (see Preparation of GST Fusion Proteins)

GST protein

GST pull-down lysis buffer, ice cold

Reagents for SDS-Polyacrylamide Gel Electrophoresis of Proteins (see Step 12)

2X SDS gel-loading buffer

Tris-Cl (50 mM, pH 8.0) containing 20 mM reduced glutathione (optional; for Step 11 only)

Equipment

Equipment for SDS-Polyacrylamide Gel Electrophoresis of Proteins (see Step 12)

Gel dryer (optional; see Step 13)

Glutathione-Sepharose beads (store at 4°C; do not freeze)

Beads are often supplied by commercial vendors in solutions containing alcohols. It is important to wash the beads thoroughly in GST pull-down lysis buffer and to generate a 50/50 slurry of beads in GST pull-down lysis buffer prior to use.

Microcentrifuge, precooled to 4°C

Microcentrifuge tubes, 0.5 mL (optional; for Steps 11.vi-11.ix only) and 1.5 mL

Needle, small bore, sterile (optional; for Steps 11.vi-11.ix only)

Rotator for end-over-end mixing

Water bath, boiling (optional; see Step 10)

X-ray film (optional; see Step 13)


METHOD


1. Incubate the cell lysate with 50 µL of glutathione-Sepharose beads (50/50 slurry in lysis buffer) and 25 µg of GST (NOT the GST fusion probe protein) for 2 h at 4°C with end-over-end mixing. Allow enough volume in the tube to permit liberal mixing; 500 µl to 1 mL is a good starting point.
This step is designed to preclear from the lysate proteins that interact nonspecifically with the GST moiety or with the beads alone. If the interaction will be detected primarily with antibodies directed to a candidate interacting protein, it is not absolutely necessary to preclear the lysates with GST or glutathione-Sepharose beads. However, when 35S-labeled cell lysates are used to identify novel protein-protein interactions, these steps can help to reduce background.
When detecting the interacting protein with antibodies to that protein, it is important to include "GST + beads" and "beads alone" controls.


2. Centrifuge at 13,000 rpm for 10 sec at 4°C in a microcentrifuge.


3. Transfer the supernatant (precleared cell lysate) to a fresh tube.


4. Set up two tubes containing equal amounts of the precleared cell lysate.

i. Add 50 µl of glutathione-Sepharose beads (50/50 slurry in lysis buffer) to each tube.


ii. Then add GST protein to one tube and the GST fusion probe to the other (~5-10 µg each).
The amount of protein added should be equimolar in the two reactions (i.e., the final molar concentration of GST should be the same as that of the GST probe protein).

5. Incubate the tubes for 2 h at 4°C with end-over-end mixing.


6. Centrifuge the samples at 13,000 rpm for 10 sec at 4°C in a microcentrifuge.


7. Transfer the supernatants to fresh microcentrifuge tubes and reserve them for SDS-PAGE (see Troubleshooting).


8. Wash the beads four times with 1 mL of ice-cold lysis buffer. Discard the washes.


9. At this point, proteins bound to the probe protein must be dissociated for analysis. Use either the boiling method (the more popular choice; see Step 10) or one of the elution methods described in Step 11.
The disadvantage of elution (Step 11) is that if multiple elutions are necessary, the final sample volume may be large. In these cases, it may be impossible to load more than a small percentage of the sample on an SDS-polyacrylamide gel for analysis. This is the reason most researchers choose instead to boil complexes off the beads in SDS sample buffer (Step 10).


10. If the samples are to be boiled off the beads, do as follows:

i. Add an equal volume of 2X SDS gel-loading buffer to the beads.
It is important to include a "glutathione-Sepharose beads only" control. Proteins bound nonspecifically to the beads can appear as bound to the fusion protein, even in comparison to GST alone.


ii. Boil for 5 min in a water bath.
Samples are now ready for SDS-PAGE (Step 12).

11. If the samples are to be eluted from the beads, perform one of the following options:
Option I:

i. Add 50 µl of 20 mM reduced glutathione in 50 mM Tris-Cl (pH 8.0).


ii. Flick the tube to mix the contents, and incubate the samples at room temperature for 5 min.


iii. Centrifuge the samples at 13,000 rpm for 10 sec at room temperature in a microcentrifuge.


iv. Transfer the supernatant containing the eluted protein to a new microcentrifuge tube.


v. If elution is incomplete, repeat Steps 11.i-11.iv.

Option II:

vi. Add 50 µl of 20 mM reduced glutathione in 50 mM Tris-Cl (pH 8.0).


vii. Flick the tube to mix the contents, and incubate the samples at room temperature for 5 min.


viii. Transfer the mixture to a fresh 0.5-mL tube. Carefully, using a sterile, small-bore needle, poke a hole in the bottom of the 0.5-mL tube. Place it inside a 1.5-mL microcentrifuge tube.


ix. Centrifuge the tubes together at 1000g in a microcentrifuge for 2 min at room temperature.
The eluted proteins will be deposited in the 1.5-mL tube.

12. Analyze as much of the sample as possible by SDS-PAGE (see SDS-Polyacrylamide Gel Electrophoresis of Proteins).


13. Detect the proteins. The method of detection will depend on the experimental goal:

i. If the goal is to detect the 35S-labeled proteins associated with the fusion protein after SDS-PAGE, dry the gel on a gel dryer and expose it to X-ray film.


ii. If the goal is to detect specific partners after SDS-PAGE, transfer the proteins to a membrane and subject them to Western analysis.


iii. If the goal is to determine the size and abundance of proteins associated with the fusion protein from a nonradioactive lysate, subsequent to SDS-PAGE, stain the gel with Coomassie blue or silver stain.


TROUBLESHOOTING

Problem: Conditions for pull-down are not optimal

[Step 13]

Solution: Analyze aliquots of equal percentage volumes (e.g., 1%) from each of the following fractions generated during the protocol:

Total cell lysate (Step 1)

Beads prior to elution (Step 9)

Eluate (Steps 10 or 11)

Beads post-elution (Steps 10 or 11)

Supernatant saved at Step 7

"Beads + GST" eluate (if applicable; Steps 10 or 11)

"Beads alone" eluate (if applicable; Steps 10 or 11)

After these samples are collected at the indicated steps, add an appropriate volume of SDS-PAGE sample buffer. Snap-freeze the samples in a dry-ice ethanol bath for analysis by SDS-PAGE (during Step 12), and perform autoradiography, Western analysis, or Coomassie staining as appropriate (Step 13).

Results from this gel will show:


The prevalence of the novel interactor in the total cell lysate (aliquot from Step 1).


How much of the interactor is bound to the GST fusion protein (aliquot from Step 9).


How much GST fusion protein + associated proteins were eluted from the beads (aliquot of eluate from Steps 10 or 11).


What fraction of the interactors remained bound (beads remaining from Steps 10 or 11).


How much of the interacting protein was depleted from the total cell lysate (aliquot from supernatant at Step 7).


Problem: Low signal

[Step 13]

Solution: If an interaction yields a low signal even though the interactor is abundant in the total cell lysate, this may indicate that the binding conditions are not optimal. A change in salt and detergent concentrations, in addition to increasing the time allowed for association, may improve the binding. A poor signal can also be caused by inefficient elution; i.e., the complex being retained on the glutathione-Sepharose beads. This can be determined by SDS-PAGE comparison of the eluate versus the "beads post-elution" fraction (Steps 10 or 11). If this proves to be the problem, it may be remedied by pooling multiple elutions or increasing the time for elution. Releasing the interactors by boiling in sample buffer (Step 10), if appropriate, would be applicable in this case.

Problem: Nonspecific background

[Step 13]

Solution: Preclearing a lysate with GST, or beads alone, can help to minimize nonspecific interactions (see Step 1). Titrating the amount of lysate added and increasing the stringency of the binding and wash conditions can also reduce background.


DISCUSSION

When analyzing an interaction by Western blot (e.g., analyzing a predicted interaction for which there are available antibodies), it is important to reprobe the membrane with anti-GST antibodies after probing for the candidate interactor. This will determine whether all samples were incubated with the same amount of GST fusion protein, and it will help to determine whether the fusion protein is undergoing degradation while incubated with the cell lysate. If the interactors are radio- or biotin-labeled, one should also confirm equal amounts of GST fusion and GST proteins by Western blot or Coomassie blue staining.



REFERENCES

Dignam, J.D., Lebowitz, R.M., and Roeder, R.G. 1982. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11: 1475–1489.

Orlinick, J.R. and Chao, M.V. 1996. Interactions of the cellular polypeptides with the cytoplasmic domain of the mouse Fas antigen. J. Biol. Chem. 271: 8627–8632.[Abstract/Free Full Text]

Spector, D.L., Goldman, R.D., and Leinwand, L.A. 1998. Cells: A laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

查看更多来自冷泉港实验室的Protocol:

CSH protocols之蛋白阵列试验

冷泉港 Protocols——权威的实验方案

CSH protocols——GST融合蛋白的表达纯化

2009年2月3日星期二

A Comeback for Lamarckian Evolution?

The effects of an animal's environment during adolescence can be passed down to future offspring, according to two new studies. If applicable to humans, the research, done on rodents, suggests that the impact of both childhood education and early abuse could span generations. The findings provide support for a 200-year-old theory of evolution that has been largely dismissed: Lamarckian evolution, which states that acquired characteristics can be passed on to offspring.

"The results are extremely surprising and unexpected," says Li-Huei Tsai, a neuroscientist at MIT who was not involved in the research. Indeed, one of the studies found that a boost in the brain's ability to rewire itself and a corresponding improvement in memory could be passed on. "This study is probably the first study to show there are transgenerational effects not only on behavior but on brain plasticity."

In recent years, scientists have discovered that epigenetic changes--heritable changes that do not alter the sequence of DNA itself--play a major role in development, allowing genetically identical cells to develop different characteristics; epigenetic changes also play a role in cancer and other diseases. (The definition of epigenetics is somewhat variable, with some scientists limiting the term to refer to specific molecular mechanisms that alter gene expression.) Most epigenetic studies have been limited to a cellular context or have looked at the epigenetic effects of drugs or diet in utero. These two new studies are unique in that the environmental change that triggers the effect--enrichment or early abuse--occurs before pregnancy. "Give mothers chemicals, and it can affect offspring and the next generation," says Larry Feig, a neuroscientist at Tufts University School of Medicine, in Boston, who oversaw part of the research. "In this case, [the environmental change] happened way before the mice were even fertile."

In Feig's study, mice genetically engineered to have memory problems were raised in an enriched environment--given toys, exercise, and social interaction--for two weeks during adolescence. The animals' memory improved--an unsurprising finding, given that enrichment has been previously shown to boost brain function. The mice were then returned to normal conditions, where they grew up and had offspring. This next generation of mice also had better memory, despite having the genetic defect and never having been exposed to the enriched environment.

The researchers also looked at a molecular correlate of memory called long-term potentiation, or LTP, a mechanism that strengthens connections between neurons. Environmental enrichment fixed faulty LTP in mice with the genetic defect; the fixed LTP was then passed on to their offspring. The findings held true even when pups were raised by memory-deficient mice that had never had the benefits of toys and social interaction. "When you look at offspring, they still have the defect in the protein, but they also have normal LTP," says Feig. The findings were published today in theJournal of Neuroscience.

"If the findings can be conveyed to human, it means that girls' education is important not just to their generation but to the next one," says Moshe Szyf of McGill University, in Montreal, who was not involved in the research.

In a second study, researchers found that rats raised by stressed mothers that neglected and physically abused their offspring showed specific epigenetic modifications to their DNA. The abused mice grew up to be poor mothers, and appeared to pass down these changes to their offspring.

Previous research has shown that bad rat mothering can be passed down through this kind of DNA modification--but those changes are thought to be triggered specifically by maternal behavior. In the new study, researchers also had healthy mothers raise the offspring of stressed mothers, and found that the problems were only partially fixed. That suggests that the changes "were not due to their neonatal experience," says says David Sweatt, a neuroscientist at the University of Alabama at Birmingham, who oversaw the study. "It was something that was already there when they were born." The research was published online last month in Biological Psychiatry.

The results of both studies are likely to be controversial, perhaps resurrecting a centuries-old debate. "It's very provocative," says Lisa Monteggia, a neuroscientist at the University of Texas Southwestern Medical Center, in Dallas. "It goes back to two schools of thought: Lamarck versus Darwin."

In contrast to natural selection, in which organisms that are born well adapted to their environment survive and reproduce, passing down those successful traits, Lamarckian evolution suggests that animals can develop adaptive traits, such as better memory, during their lifetimes, and pass on those traits to their offspring. The latter theory was largely abandoned as Darwin's, and later Mendel's, theories took hold. But the concept of Lamarckian inheritance has made a comeback in recent years, as scientists learn more about epigenetics.

"I didn't set out to come up with findings that support neo-Lamarckian inheritance," says Sweatt. "But the research now makes it more plausible that these things may be real and may be based in molecular mechanisms."

Feig, on the other hand, argues that while the findings are "a Lamarckian kind of phenomenon it's still Darwinian, because the changes don't last forever." In Feig's study, the offspring of enriched mice lost their memory benefits after a few months.

Sweatt and others say that this type of inheritance may in fact be much more common than expected. Improving technologies are now providing a broader look at the epigenetic changes linked to environment and behavior. Scientists are starting to use DNA microarrays, which over the past few years have become widely employed in genetic studies of disease, to look at one specific type of change, known as DNA methylation. "The changes we see are not limited to a small number of genes," says Szyf, who is using the technology to study epigenetics and cancer. "Whole circuitries are changed."

DNA sequencing, which is rapidly dropping in price, can also be used to study DNA methylation. But epigenetics studies require high-volume sequencing, which has been prohibitively expensive. "In contrast to the genome, every epigenome is different in different types of cells," says Sweatt. "A human epigenome project would be the equivalent of 250 human genomes, because there are at least 250 cell types in the body." Cheap sequencing may soon make that type of study possible, he says.

The actual mechanism underlying these patterns of inheritance is somewhat mystifying to scientists. Feig theorizes that environmental enrichment triggers a long-lasting hormonal change: when the animal becomes pregnant, the hormone would somehow modify the DNA of the fetus, ultimately causing it to have improved memory and LTP as an adolescent. However, he cautions, there is no direct evidence of this, and no specific evidence that the behaviors are transmitted through epigenetic mechanisms.

原文:http://www.technologyreview.com/biomedicine/22061/page1/