Statistics - anyone able to help with some uni work?

Spuddy

Spuddy

Spuddy

Associate
Joined
24 Mar 2012
Posts
1,744
Location
Im In Me Mums Car
I have some work at uni that requires me to do some statistics, problem is I have absolutely zero idea what Im doing and my personal tutor is rather useless too and similarly has no idea what he's doing. So hoping the good folks over here at OCUK may be able to help. Hopefully theres a statistician amongst you lot!

The statistics is for my Biomedical degree (final year project) and I've tackled the actual biology side of things just fine, its just this statistics holding me back from progressing. Im eager to get back in the labs and wrap things up.

So if anyone can help please let me know and I can drop you an email with the excel file containing the data.

Many thanks!
 
You are final year of a biomedical degree and don't have a clue about statistics?

What exactly are you struggling with?

Surely this is something fairly basic and important to get your head round when doing any research work or lab work. spend 20 minutes of reading up and you will be a stats pro. Even college level statistics only really involves a primary school level of maths.
 
Last edited:
Go get some statistics books from the library and start learning...there isn't enough posts in this forum that can teach you stats to pass your degree.
 
What raymond said, its very dependant on what you want from your data as to how you'd go about analysing it, which means if we were to help you then we'd basically be doing it for you.

Univeristy doesnt just mean lectures and exams, it means being able to read around the subject and to learn anything you need to know to get the job done, as my dad would put it it's more of a qualification in problem solving than actual knowledge.

The short of it is, as i'm sure we'll be seeing someone else posting soon; do your own damn homework
 
Avenged7Fold said:
You are final year of a biomedical degree and don't have a clue about statistics?

What exactly are you struggling with?

Surely this is something fairly basic and important to get your head round when doing any research work or lab work. spend 20 minutes of reading up and you will be a stats pro. Even college level statistics only really involves a primary school level of maths.
Click to expand...


Statistics isn't something we really go through as part of our degree, its 99% molecular biology and chemistry. Which I'm 1st class honours student in, its just my brain doesn't seem to be wired for mathematical statistics.

Im unsure on which test(s) to run. I already have P-values for my data so T-test doesn't seem relevant in this scenario. The data is basically comparing two lines of oesophageal cancer cells (radio/chemo resitant vs radio/chemo sensitive).

I am trying to get my head round it, thats why I've come here for help. I don't want someone to do the work for me, Im only asking for help.


adolf hamster said:
Univeristy doesnt just mean lectures and exams, it means being able to read around the subject and to learn anything you need to know to get the job done, as my dad would put it it's more of a qualification in problem solving than actual knowledge.

The short of it is, as i'm sure we'll be seeing someone else posting soon; do your own damn homework
Click to expand...


Yes thanks Im well aware of what University entails at this stage, Im just struggling to get my head around statistics - its not exactly the primary focus of the degree and wasn't expecting to encounter something like this. I guess you either just get something or you don't and it requires extra help.

I am doing my own 'damn homework', Im asking for help and guidance not someone else to do it for me. I don't see the/your issue? How am I supposed to learn if I don't seek the knowledge/info from elsewhere to teach me first.
 
Guys, give the guy a break.

What are you trying to do with your data?

At the very least, people can help point you in the right direction.
 
Thanks guys!

Basically, I've been given a set of data for Oesophageal cancer cells. I need to compare the lines of cells to find commonalities in differentially expressed microRNA between the different groups (chemo resistant and chemo sensitive - with stem-cell variants of each group). To try and find some correlation between certain microRNA(s) being responsible for either the sensitivity or resistance of the cell. ... If that makes sense?

A quick snapshot of part of the data looks like this, I believe its pointless running a T-test as the P-values have been provided?

Column on the left (miR) is the miRNA. OE33 is the oesophageal cell lines. A-ve = non-stemcell. A+ve= stemcell


image hosting
 
Last edited:
AHarvey said:
Daft question, could you take the data to a maths teacher/lecturer there and ask them to give you guidance?
Click to expand...

I guess I could ask yeah, its just tracking them down and finding one which would be the problem. Thanks for the suggestion.


Nix said:
Bare in mind that I'm absolutely no maths expert, but is this what you need?

http://www.investopedia.com/terms/a/anova.asp
https://en.wikipedia.org/wiki/Analysis_of_variance
Click to expand...


I have no idea if an anova is what I need at this point but I'll take a look, thank you very much.

Man, I hate stats!
 
If you have a p-value you just need a level of significance to see if the null hypothesis is any good. If you have 1% significance then any value for which p<=0.01 rejects the null hypothesis.
 
Last edited:
Hmm, I did molecular biology at uni and now do stats, not sure if I'll get time to properly look however
 
Spuddy said:
I have some work at uni that requires me to do some statistics, problem is I have absolutely zero idea what Im doing and my personal tutor is rather useless too and similarly has no idea what he's doing. So hoping the good folks over here at OCUK may be able to help. Hopefully theres a statistician amongst you lot!

The statistics is for my Biomedical degree (final year project) and I've tackled the actual biology side of things just fine, its just this statistics holding me back from progressing. Im eager to get back in the labs and wrap things up.

So if anyone can help please let me know and I can drop you an email with the excel file containing the data.

Many thanks!
Click to expand...


You've failed at the first hurdle. Your entire experiential design should have been designed around the statistical tests available and expected power of the effect size.
 
Spuddy said:
Thanks guys!

Basically, I've been given a set of data for Oesophageal cancer cells. I need to compare the lines of cells to find commonalities in differentially expressed microRNA between the different groups (chemo resistant and chemo sensitive - with stem-cell variants of each group). To try and find some correlation between certain microRNA(s) being responsible for either the sensitivity or resistance of the cell. ... If that makes sense?

A quick snapshot of part of the data looks like this, I believe its pointless running a T-test as the P-values have been provided?

Column on the left (miR) is the miRNA. OE33 is the oesophageal cell lines. A-ve = non-stemcell. A+ve= stemcell


image hosting
Click to expand...





Can you try and explain the data again but ignoring the underlying microbiological nomenclature? E.g, can you phrase in terms of Variables A, B and C, e.g. you want to know how variables A anD B correlate with C? Particularly, describe what the data is n each column and what you hope to show. Remember, an important part of getting a degree is explaining highly technical work to people that are no in your field, so this is useful practice.

You have 4 columns, the first looks like a description of a particular experiment. You then have 2 columns of data, how do they relate? You then have some probability values, how are these values calculated, are there any corrections done to these p-values?


Are you simply trying to find a correlation between the numbers in the 2nd and 3rd column?
 
You must log in or register to reply here.
Back
Top Bottom