This assignment is to be done either individually or in pairs. Do not show your code to any other group and do not look at any other group's code. Do not put your code in a public directory or otherwise make it public. However, you may get help from the TAs or the instructor. You are encouraged to use the RPILMS Discussions page to post problems so that other students can also answer/see the answers.

Sequence Alignment

How to determine the similarity between two DNA sequences, for example, ACATCTCTC and AGTTTC? We can align them together and accumulate the differences between pairs of characters. For example, the following is an alignment of the two sequences.

A C A T C T C T C
A G T T T C - - -
0 1 1 0 1 1 2 2 2

The gap character '-' is used to make the sequences equal in length. The difference between the same characters is 0. The difference between different characters is 1, and the difference between a character and a gap is 2. The difference of the alignment is the sum of the character distances: 0+1+1+0+1+1+2+2+2 = 10. Two sequences can be aligned in many ways. The following is another alignment of the sequences whose difference is 0+1+2+0+1+0+0+2+2 = 8.

A C A T C T C T C
A G - T T T C - -
0 1 2 0 1 0 0 2 2

Define the distance of two sequences as the minimum difference of an alignment of them, which is a good indication of their similarity. We can compute the distance using dynamic programming. Given two sequences s1 = a1,a2,a3,...,am and s2 = b1,b2,b3,...,bn, define distance(i,j) (0≤i≤m, 0≤j≤n) as the distance between sub-sequences a1,a2,...,ai and b1,b2,...,bj. Then distance(i,0) = 2i and distance(0,j) = 2j, when one sub-sequence is empty. distance(m,n) is the distance between s1 and s2. In an optimal (minimum difference) alignment of sub-sequences a1,a2,...,ai and b1,b2,...,bj, there are three possible cases:

1. ai is aligned with a gap, and the sub-alignment of a1,a2,...,ai-1 and b1,b2,...,bj is optimal. distance(i,j) = distance(i-1,j) + 2.
2. bj is aligned with a gap, and the sub-alignment of a1,a2,...,ai and b1,b2,...,bj-1 is optimal. distance(i,j) = distance(i,j-1) + 2.
3. ai is aligned with bj, and the sub-alignment of a1,a2,...,ai-1 and b1,b2,...,bj-1 is optimal. distance(i,j) = distance(i-1,j-1) + (ai==bj ? 0 : 1).

Which indicate that distance(i,j) = min(distance(i-1,j) + 2, distance(i,j-1) + 2, distance(i-1,j-1) + (ai==bj ? 0 : 1)). We can store all the distances in an (m+1)×(n+1) matrix M, distance(i,j) = M[i][j], and compute M[m][n] using the following pseudocode.

M[i][0] = 2 * i
M[0][j] = 2 * j
for i = 1 to m
    for j = 1 to n
        M[i][j] = min(M[i-1][j] + 2, M[i][j-1] + 2, M[i-1][j-1] + (ai==bj ? 0 : 1))
    endfor
endfor

Part 1 (80%). Concurrent programming.

First, read the comprehensive example of SALSA.

You are given a file sequences.txt containing N (= 100) random DNA sequences, each sequence on a line. Your task is to find the three most similar pairs of sequences. You can assume that the length of a sequence does not exceed 100. Write a concurrent sequence alignment program in SALSA. There should be M worker actors and one coordinator actor. The worker actors compute the distances of pairs of sequences and the coordinator actor gives the pairs to the workers and merges the results. Each worker actor is responsible for N(N-1)/2M pairs of sequences. All the actors run locally in a single theater in Part 1. Your program is required to take two arguments: sequences.txt and the number of worker actors. The command to execute the program should look like the following:

$ java /* your program */ sequences.txt 10

Where /* your program */ is the name of your program and 10 is the number of worker actors. Your coordinator actor should output the top pairs and the total running time:

sequences: 100, workers: 10, theaters: 1
most similar pairs:
distance(GGGACGATAAAGAAGGAGATTGGATTGCCAAAGCAACATTGTGAGCAAATAAACAGGCATATGGTCATACG
GTCGCATCTCCATGGGATG, TGCCCCCTAGAAAAGGGATATACTGAAACGAAACTATCATGCTACCGAACTCTCTAGGG
ACCATGGTCTAGGCAGCCTGTGTCATATATTT) = 52
distance(AAGTGCCAGTATATCCTACAGAGCGACATGGACCAGAAGCTGACGGCCTCGCTATAAAACATTATCATGTC
CAGATCTCGCGAAGATGCGACTGC, GATGCGAACAACCGTTAGCCACGTTCTGGAACATATACCGAGCGCAGGCCAACC
ATTCGGCAATTCGCCCGGAGCTGCCCAAAAAGCGATGG) = 53
distance(GATGCGAACAACCGTTAGCCACGTTCTGGAACATATACCGAGCGCAGGCCAACCATTCGGCAATTCGCCCG
GAGCTGCCCAAAAAGCGATGG, GTGGCGTAAACCCGTTTGGATGCGAGGCCTGAAGGCCCTTTGTCGCACAACCCCTAG
ACACGTGTCTGCACGAGACGGCCGGAATAGCTAGT) = 53
total running time: ....ms

Part 2 (20%). Distributed programming.

Write a distributed version of the program in SALSA based on Part 1. That is, you should use multiple theaters and distribute worker actors on each theater. In addition to the arguments used in Part 1, your program must accept another argument to specify theaters and a nameserver as follows:

$ java /* your program */ sequences.txt 10 theaters.txt

The theaters.txt is a text file, the first line of which is the location of the nameserver and the rest are locations of theaters. An example of theaters.txt is here.

Extra Credit (up to 25% bonus).

1. Add a load balancing capability to your program. Instead of creating the same number of worker actors on each theater, create all workers on a single theater. Then, have other theaters steal some worker actors from the theater with many worker actors. Include the test code and an explanation of the results in your submission.
2. See the professor if you have ideas for other extensions to this assignment and would like extra credit for implementing them.

Due Date:

Grading: The assignment will be graded mostly on correctness, but code clarity / readability will also be a factor (comment, comment, comment!). See the professor or TAs, if you have ideas for other extensions for this assignment and would like extra credit for implementing them.

Submission Requirements: Please submit a ZIP file with your code, including a README file. In the README file, place the names of each group member (up to two). Your README file should also have a list of specific features / bugs in your solution. Your ZIP file should be named with your RPILMS user name(s) as the filename, either userid1.zip or userid1_userid2.zip. Only submit one assignment per pair via RPILMS.