Ecological Immunology - Science topic

M. A. A. Al- Fatlawi wow this is great! I was struggling to find any papers on the subject. Do you happen to know of any more direct studies linking erythrocyte infection and peptide presentation on mHC class I in poultry or other avian systems?

Thanks for your help

Any standard flowcytometer instrument use for humans should do it. Sysmex, advia, coulter is most common. Take tke samples to your local hospital and talk eith the lab. See https://pdfs.semanticscholar.org/2ceb/bbf8230a08b64b2e4a345a628642e4eaaabe.pdf

I am actually interested in HLA-B, C and DBR1

There was a paper recently published in 2015 by the American Journal of Clinical Pathology that demonstrated complement factors are stable in citrated plasma when stored up to 6 years at -80˚C.  We have good reproducibility in our assays with NHP serum in storage for 1 year at -80˚C.  As long as the serum you collected wasn't sitting at RT for more than 3 hours prior to storage, you should have good hemolytic or microbe killing assay results.  The article I mentioned gave a comprehensive investigation in preanalytical effects on complement biomarker assays.  It's not the end-all guideline for processing and storing serum/plasma for complement activity preservation, but it clarifies the impact of long-term storage on complement studies and I highly suggest giving it a read for your future experiments.

You might also consider a microbial killing assay modified for small volumes of plasma to give a functional measure of immune defense. You can chose the model pathogen based on the aspects of immune defense pertinent to your system. The two papers below describe the technique and how to validate it for your host species and the laboratory pathogen (E. coli, S. aureus, etc). You perform the MKA on a spectrometer and it only requires 3-4 microliters of plasma.

Dear Agnieszka,

There are no primers that could amplify all V regions. You could get a couple for the C (specific for each mouse strain), but not for the V. You can design several, I tried myself and design almost 30. The problem is you run out of your sample pretty fast if it is a tiny population. Another option is to extract RNA and perform a 3' RACE using C primers. It is a technique that allows amplification of sequences when 3' region is unknown. You can clone and sequence the product you get. I hope this helps. Good luck!

Some acronyms might help here:

LM General linear model -- linear model (usuall assumed normal theory least squares = maximum likelihood)

GLM -- General linear model (not Generalized) is the family of models of the form Y=XB+E and includes regression, anova, ancova and even repeated measures designs for continuous dependent variable, usually assumed normally distributed.

Generalized LInear Model (GLIM) --Extending the GLM to allow for a nonlinear relationship between the response and the predictors specified by a link function like logit for binomial etc. This also extends the GLM beyond the normal theory asumption to include a range of distributions within the exponential family--But the link(Y) is assumed to be linear in XB and the errors are assumed to be independent

MIxed models generalize the GLM to include correlated errors, fixed and random effects, nested structures and other more complex designs but with the same linearity and normal theory errors

Generalized Mixed Models extend the GLIM models to also include non-normally distributed response (ie. poisson, Binomial etc. ) and with dependent error structures--like the "mixed" models--mixed models for non-normally distributed data.

Now GAMs are a bit of a different animal altogether because there is the added complexity of "unknown" functions of the independent variables worked into the above hierarchy. This aspect of the GAM conceptualization leads to a very different frame of inference as the focus is often on the underlying functional form that is at the outset unknown.

From Wikepdia:

In statistics, a generalized additive model (GAM) is a generalized linear model in which the linear predictor depends linearly on unknown smooth functions of some predictor variables, and interest focuses on inference about these smooth functions. GAMs were originally developed by Trevor Hastie and Robert Tibshirani [1] to blend properties of generalized linear models with additive models.

I'm sure I'm not complete with this outline, but this is pretty close to how these models relate to each other.

I will have a look :) Thanks

Andy Davis at UGA has done a great deal of work on wildlife hematology. You might find this site he's put together, on avian white blood cells, useful.