The effects of UV light on bacteria growth or mortality...

[clau0329]

I would really appreciate it if someone can help me on my science project, which is: What are the effects of adding ultraviolet light on the growth or mortality of bacteria. The problem is, I need a specific bacteria to test on, and I would not know exactly which type it is without conducting the experiment first. However, this year, the amount of time we can work on a project is rather limited, so I cannot perform that option. I am planning on using the bacteria found on a doorknob, as it is a household item that carries one of the most bacteria, for my project and it would be helpful if I can recieve information on the matter. If it's not possible, well then I will still be greatful for any effort involved. And just to include further details, I'd like to know if because we always carry varied amounts and types of bacteria, will it just be more effective to grow a certain type, so as to eliminate other such factors? It worries me that I will work hard on a project, just to find out a major aspect will hurt the results. I'm sorry if I am being a bit problematical, but I cannot go to a lab facility to grow bacteria straightfowardly, so I want another option for as the type of bacteria needed and how to grow it without much cost. Thank you very much; I appreciate it if you will answer to my requests.

[clau0329]

Well, I have not gotten anywhere since my last visit, and I'm in sort of desperate need. I believe that I wrote my statement incorrectly last time. Just in case, it's: What is the effect of adding various amounts of ultraviolet light exposure on the growth and mortality of bacteria? Anyway, beacuse I am in eight grade, this is the last year that I will be required to perform a science project and I want it to be very good. Please, if anyone can, may I have some help as to what type of bacteria I should grow and how without spending much money (money is limited this year in my family)? I'd really appreciate it. Thanks.

[jenniferpaulson]

Hi. Photosynthetic bacteria (cyanobacteria) and non-photosynthetic bacteria (eubacteria) may respond differently in this experiment. You should have make sure that your of bacteria does not contain a mixture of the two. (In fact - a great experiment would be to test the two types separately!).

I'm not sure about inexpensive sources for these two types of bacteria - I'll try to look into it...

Jennifer

[jenniferpaulson]

You can buy bacterial cultures from these companies (below). They cost about $10. I think that your teacher needs to make the order.

* Presque Isle Cultures
814-833-6262
* American Type Culture Collection
703-365-2700
* Carolina Biological
800-334-5551

Maybe you could ask a local university for some bacteria. A typical bacteria that is studied is a non-pathogenic strain of E. coli. They are not photosynthetic.

If you use the bacteria on a doorknob and grow them on a petri dish, you should be able to tell the photosynthetic bacteria due to their color. They can be yellow, red, brown, green, blue-green, or black.

General info about different types of bacteria:
http://www.sirinet.net/~jgjohnso/monerans.html

Info from Science Buddies about ways to culture bacteria, etc. (look under Resources):
https://www.sciencebuddies.org/mentorin ... ?from=Home

Jennifer

[surekhamuralidharan]

Hi.

I have a suggestion that you should get the help of your current or previous science teacher for this project. This is because it is usually not allowed for a person without permission to grow bacterial cultures at home. This was given to me as advice from a science fair course I am taking right now.

For more information you can also visit
http://www.madsci.org/FAQs/micro/culture.html

Surekha

[hhemken]

How about using baker's yeast instead of bacteria? You can buy it in the baking section of most supermarkets.

If you buy packets of Fleischmann's Activated yeast, make sure they all are from the same manufacturing lot or batch, and each time you prepare a culture, do it in exactly the same way. You will thus have some highly reproducible results, and can do maybe 5 cultures for each treatment group. In the end, I bet you would have excellent statistical results.

If you can use a laboratory UV light, you will probably be able to compare at least two UV wavelengths. Maybe you can test one or two substances as UV protectors. I'm not sure how you would quantify yeast viability. Maybe just turbidity using your school's colorimeter if they have one. If you culture them in test tubes, you can put a balloon on the test tube and it will inflate with the carbon dioxide normally excreted by the yeast. You could measure balloon diameter as an indicator of yeast viability. Use a digital camera in a setup where the distance from the camera to the test tube is always the same, then perform measurements on the digital photo. Bear in mind that many types of glass absorb UV.

[clau0329]

Well, I found most of the posted replies to be helpful, so I am very grateful for those who responded ( ), but now my teacher is requiring us to make a statistical analysis on our projects ( ). I'm slightly confused because we have to choose from a t-test or chi-square to do it on. Since I am basically putting the bacteria under UV for 5, 10, and 20 minutes to see at what time of exposure the bacteria mortality will begin, I don't know if I'll need the t-test or chi-squared. At the present time, my teacher does not want to help us in our own matters, and I believe it's from her own personal time that it's being caused from, so I need someone else's help. If I use the t-test, what would I be comparing?? I'm super sorry, but I'm practically distressed. Anyway, by using x, what would be the frequency distribituion, if any? I don't think that there will be a frequency distribution predicted by chance, so I quess I'm stuck with the t-test, right? But I don't know for what samples I would be calculating the means for if I have 3 different levels of independent variables. Or is it that I need a whole entire statistical test to perform? My teacher never taught us any others, but I don't see how I may use either of these two tests. If not, I just do not realize what I am doing incorrectly. . So, if anyone else cares about this experiment, please reply and tell me if I need a new test to analyze my data. A whole heap of thanks if you can find it to help me by tomorrow or Monday.

[Lise Byrd]

What exactly are your independent variables?

I think you would be best off with a t-test. A t-test shows whether the means (averages) from two sets of data are the same or different. Based on your independent variables, you should be able to graph the mortality rates of your bacteria as straight lines. If I understand your experiment correctly, you would then use the t-test to show that the mortality rates are, indeed, different. It's hard for me to say more without knowing the precise variables.

I know you didn't ask about this, but you might try some times under 5 minutes and between your other times as well. This will give you more points to plot on a graph and give you a better indication of what the line looks like.

Sonia

[clau0329]

I want to thank you again for replying...but I have another question. Since my independent variables are the amounts of time I will be putting the bacteria under the uv for, which are 5, 10, and 20 minutes, I do not know how to set up the t-test. Where do the means come from? For example, in the null hypothesis, I think I am supposed to state, "The mean___ of ___ is not significantly different from the mean ___ of ___.", right? But, I do not understand this. I am so sorry. Believe me. I also don't get the idea of 'graphing the mortality rates of my bacteria as straight lines.' Please forgive me. If I have to calculate the means (which in truth I have no idea of where they come from) and variance for two sets of data, what will the two sets be?

[phamlinh]

Hi there,

From what I understand, you're going to expose microbes to UV light for different amounts of time and see how many of the microbes die. It's awesome that you're already thinking about statistics and how to set up your experiments.

First, let's talk about how you measure microbe death. Because you want to do the experiments at home, I strongly suggest you follow hhemken's suggestion of using Baker's yeast. If you just culture bacteria off the door, some of the strains may be technically pathogenic (i.e. dangerous) and would require a lot of approval before you perform your experiments. Additionally, you would introduce lots of variables like knowing that you have the optimum growth conditions for the different bacteria. Baker's yeast is very well defined, you can start your cultures in exactly the same way, and you can be certain that you're limiting the number of variables you're introducing into your system.

There are generally two ways to measure microbe death. Make sure to have an untreated control so that you know how the yeast would react normally. First, you can follow the suggestion posted above and monitor yeast proliferation in media. The more yeast that are growing in the media, the more cloudy the media gets. You can measure this using a spectrophotometer to get an absorbance reading. Basically, you would be measuring how cloudy/turbid the solution got, and this is directly correlated with how many bacteria are present in the liquid culture.

Alternatively, you can plate dilutions of yeast on agar plates (your teacher can help you make these) and count the number of colonies that grow. At the correct dilution, a single colony corresponds to growth from a single yeast microbe. In this way, you can quantify how many yeast are killed versus the control.

In either case, you need to have replicated samples for each condition. For example, you'd have 3 samples treated for 5 minutes, 3 samples treated for 10 minutes, 3 samples treated for 20 minutes, and 3 samples that are not treated at all. For each individual test, you need to quantify the number of yeast that are present (by measuring the absorbance, or by counting the number of colonies). Because you have replicates, you can calculate the mean for each condition. A t-test is appropriate in this case because you can compare whether the mean growth of yeast treated with UV for 5 minutes and the mean growth of untreated yeast. So you have these two values (the means), and now you want to know if they're significantly different. This is where the t-test comes in. You perform a t-test to demonstrate that the probability of the two means occuring by chance is very low. In other words, 95% of the time that you perform the experiment, you would expect the means to be close to the values you derived in your experiment.

I hope this explains your experimental design and the purpose of a t-test a little better. If it's unclear, let us know.

As for "graphing the mortaility rates as a line", that's something separate from the t-test. Now we're talking about making a correlation. For example, you may expect regardless of hte amount of time the microbes are exposed to UV light, you'll get the same amount of killing. This would be a straight line between your 5, 10, and 20 minute treatments. Let's think about a different hypothesis. What would you expect to happen to the line if longer UV light treatments kill more bacteria?

I hope this was helpful. Please post if any of this is unclear.

Best of luck,
Linh

[clau0329]

Well, I am extrmely grateful for everyone's post and for helping me on my science project. It did me a lot of good to understand all of it, and I would especially like to thank Linh. The statistical advise was necessary and I appreciate it so much. I just don't know how to thank you enough. Under all personal circumstances, I did my best and I repeat, THANK YOU! You guys are like miracle workers! Well, my gratitude for all of you is sincere, and I also wish you all the best in everything you do.
THANK YOU! (These are happy tears, by the way. )