Description
For this assignment, you will build a simple analysis program that will compute basic statistics
for a list of DNA strings. Your program should work as follows:
-The program will accept as a command line argument the name of a text file that will contain an
arbitrary list of DNA strings. (ie. ~/assign1/filename.txt) DNA strings consist of a sequence of
nucleotides (A,C,T, or G). There will be 1 string per line of the file. No guarantees on
capitalization.
-The program will then compute the sum, mean, variance, and standard deviation of the length of
the DNA strings in the list. It will also compute the relative probability of each nucleotide
(A,C,T, or G), as well as the probability of each nucleotide bigram (AA, AC, AT, AG, CA, CC,
CT, CG, etc) across the entire collection.
-The program will output the labeled results to a file called yourname.out. At the top of the file,
output your name, student id, etc.
-After printing the summary statistics to yourname.out, you will generate 1000 DNA strings
whose lengths follow a Gaussian distribution with the same mean and variance as calculated
above. The relative frequency of nucleotides will also follow the statistics calculated above.
Append the 1000 strings to the end of yourname.out.
-The program will then ask the user if they want to process another list.
-If not, the program will exit. If so, the program will prompt for the name of the next file, process
it, and append the results to the output file.
Hints
To generate the length of a string from a Gaussian distribution with mean µ and variance
σ
2
, you can use the rand() function (normalized – see RAND_MAX) to generate 2
random numbers, a and b, uniformly distributed in [0,1). Using the Box-Muller
transform it is then possible to compute a random variable C, such that:
C = sqrt(-2 ln (a)) * cos(2πb)
Here C is a standard Gaussian with mean 0 and variance 1. You can then convert to a
normal random variable D with mean µ and variance σ
2 as follows:
D = σC + µ
Note that here we use the standard deviation, σ, which as you know from basic statistics
is simply the square root of variance.
The Rules
– You may NOT use any non-primitive data structures to do the math. (No arrays,
Vectors, Lists, etc) Just use individual primitive variables (int, double, etc) and std
strings. Hopefully this will convince you that data structures make programs more
efficient and easier to write. (Though I suspect you know this already…) Of course, to
do the file processing you may use any of the C++ IO classes.
– For this assignment, you must work individually. You may work in pairs in future
assignments, but I want you all to write some code on your own first.
– Develop using any IDE you want, but make sure your code runs correctly with g++.
Make sure to provide a Makefile so I can build your code easily. (See “Assignments” for
a sample Makefile.)
– Feel free to use whatever textbooks or Internet sites you want to refresh your memory
with C++ IO operations, just cite them in a README file turned in with your code. All
code you write, of course, must be your own.
Due Date
This assignment is due at 11:59pm on 9-13-2017. Submit it using Blackboard following the
submission instructions available under “Course Documents”. We’ll spend a few minutes
discussing your experience with this assignment in class.
Grading
Grades will be based on correctness, adherence to the guidelines, and code quality (including the
presence of meaningful comments). An elegant, OO solution will receive much more credit than
procedural spaghetti code. I assume you are familiar with the standard style guide for C++,
which you should follow. (See the course page on Blackboard for a C++ style guide.)
