Although there are noted dissimilarities between mouse models and human outcomes, the number of mouse models that are highly useful is large. In terms of my own area, mouse models of antiepileptic drug action, anxiolytic drug action, neuropathic pain therapy and many other similar disease treatments are highly evolved, widely published and extremely useful, not only for predicting human efficacy but also for predicting the unbound plasma drug concentrations likely to be therapeutic.

Although it always adds confidence to use multiple models and multiple species, in my experience, many mouse models are extremely useful.

Mouse models have long been a very important source of knowledge about human diseases. This is also the situation today. However, we must recognize the challenges we have when we choose animal models, and carefully choose the most relevant models .

It helps if you know that the biological system you are researching is also present in the mouse model you are using.

Otherwise, rodents are decent in vivo systems to be employed in science research, especially for pharmaceutical research. Like human, they have gastrointestinal tracts and double circulatory system. Although there are some differences between quadrupeds and bipeds (which could be significant in some cases) and baseline metabolic rates, rodents are relatively inexpensive models that mimic human system, making them suitable for in vivo experiments to answer some important research questions.

For obvious reasons, we cannot use human in most basic science research. There are many different animal models available for selection, and they all have their own distinct flaws. Therefore, interpretation is key.

Ask a specific research question, and then choose the most appropriate model or a combination of models to answer that question. This is the approach I am most comfortable with.

I once developed an animal tumor model not in mouse but in rat; although many animal models may use the identical name of a human disease, there are quite often big differences in histology: many tumors in mice and rats may share an identical nomenclature, but the histology and biological behaviour are not really comparable to their suggested human counter-part. Therefore, it was quite surprising to my bosses, when the tumors I had induced where histologically indistinguishable from their human/children counter-parts: not only the bipotential differentiation potential with glial and neuronal markers, but also the formation of neuroblastic rosettes and migration potential showed all characteristic signs of PNETs (primitive neuro-ectodermal tumors) and medulloblastoma (some childhood brain tumor). With some of the etiologic background I postulated I even started (without really telling my bosses) in my free time (i.e. a week) the analysis of human counterparts... and found the worlds first p53 mutation in this type of cancer (primary tumor tissue, not cell-line), but at a low frequency. 20 years later, Germanys cancer research appears to focus on this subtype of tumors with this type of mutation. If I remember right, there will be spent over 100 million EUR in a decade - although the tumor is not that common.

Article A model for primitive neuroectodermal tumors in transgenic n...

PS: this is my first detection of a p53 mutation in a human tissue of medulloblastoma (together with a similar finding in an oligodendroglioma)

Article p53 Mutations in Nonastrocytic Human Brain Tumors

They are a terrific source of knowledge about mice, but caution should be taken extrapolating patterns to other species.

Since mouse and human share almost any gene, a lot of diseases could be revealed just by knocking out so many genes. Unfortunately, it is expensive to keep the many strains of mice, lacking a single gene or a combination of certain genes alive, or even store their lines as frozen cells, but I think Europe works together to have such a wonderful bank of knowledge.

Mouse models reliable sources of knowledge:

Yes - Mouse are reliable source of knowledge at preliminary research.

No- incase where you are perform advance study related to your work.

Any experimental work are very important to explore inital reslult. Then it is recomended that you go other advance model and cell based assay.

Arturo Rosenblueth wrote in 1945: "the best model for a cat is another, or preferably the same cat".

Mice are great models for basic scientific questions, but many scientists would argue they are not good as models for translational science. Therefore, if quick impact/application to human health is desired then higher mammalian species like primates or human cell lines (such as induced pluripotent stem cells) are a much better model. See: http://www.csee.wvu.edu/~xinl/papers/role_model.PDF

Agree with charles Taylor, Robert Eibl

Mice models can not be regarded as surrogate to the human outcomes. But the inclination revealed by these models is reliable. Inferences are important, as we have many publications in which the magnitude of the outcome mismatches with human subjects, but the trend is analogous. After all, we donot have any reliable option other than rodents. Many researchers promote cell line studies, but cell lines can not account for/mimic for/the organized tissues/organs in live form. Research is based on inferences and should be judiciously extrapolated.