public MainForm()
{
InitializeComponent();
statusMessage.Text = "Loading Database...";
// load database here
try
{
m_sequences = loadFile("../../" + GENOME_FILE);
}
catch (FileNotFoundException e)
{
try // Failed, try one level down...
{
m_sequences = loadFile("../" + GENOME_FILE);
}
catch (FileNotFoundException e2)
{
// Failed, try same folder
m_sequences = loadFile(GENOME_FILE);
}
}
m_resultTable = new ResultTable(this.dataGridViewResults, NUMBER_OF_SEQUENCES);
statusMessage.Text = "Loaded Database.";
}
public MainForm()
{
InitializeComponent();
statusMessage.Text = "Loading Database...";
// load database here
try
{
m_sequences = loadFile("../../" + GENOME_FILE);
}
catch (FileNotFoundException e)
{
try // Failed, try one level down...
{
m_sequences = loadFile("../" + GENOME_FILE);
}
catch (FileNotFoundException e2)
{
// Failed, try same folder
m_sequences = loadFile(GENOME_FILE);
}
}
m_resultTable = new ResultTable(this.dataGridViewResults, NUMBER_OF_SEQUENCES);
statusMessage.Text = "Loaded Database.";
}
/// <summary>
/// this is the function you implement.
/// </summary>
/// <param name="sequenceA">the first sequence</param>
/// <param name="sequenceB">the second sequence, may have length not equal to the length of the first seq.</param>
/// <param name="resultTableSoFar">the table of alignment results that has been generated so far using pair-wise alignment</param>
/// <param name="rowInTable">this particular alignment problem will occupy a cell in this row the result table.</param>
/// <param name="columnInTable">this particular alignment will occupy a cell in this column of the result table.</param>
/// <returns>the alignment score for sequenceA and sequenceB. The calling function places the result in entry rowInTable,columnInTable
/// of the ResultTable</returns>
public int Score(GeneSequence sequenceA, GeneSequence sequenceB, ResultTable resultTableSoFar, int rowInTable, int columnInTable)
{
initialize(sequenceA, sequenceB);
for (int i = 1; i < height; i++)
{
for (int j = 1; j < width; j++)
{
int diagCellCost = 0;
if (X[j - 1] == Y[i - 1])
diagCellCost = CharsMatchCost;
else
diagCellCost = SubstitutionCost;
double topCell = prev[j] + InsertDeleteCost;
double leftCell = results[j - 1] + InsertDeleteCost;
double diagCell = prev[j - 1] + diagCellCost;
double min = Math.Min(topCell, Math.Min(diagCell, leftCell));
results[j] = min;
}
currentRow++;
SwapArrays();
}
return (int)prev[width - 1];
}
/// <summary>
/// This is the function you implement.
/// </summary>
/// <param name="sequenceA">the first sequence</param>
/// <param name="sequenceB">the second sequence, may have length not equal to the length of the first seq.</param>
/// <param name="resultTableSoFar">the table of alignment results that has been generated so far using pair-wise alignment</param>
/// <param name="rowInTable">this particular alignment problem will occupy a cell in this row the result table.</param>
/// <param name="columnInTable">this particular alignment will occupy a cell in this column of the result table.</param>
/// <returns>the alignment score for sequenceA and sequenceB. The calling function places the result in entry rowInTable,columnInTable
/// of the ResultTable</returns>
public int Align(GeneSequence sequenceA, GeneSequence sequenceB, ResultTable resultTableSoFar, int rowInTable, int columnInTable)
{
if((columnInTable - rowInTable) < 0)
{
return 0;
}
Grid grid = new Grid(sequenceA.Sequence, sequenceB.Sequence, true, MaxCharactersToAlign);
return grid.CalculateScoreSolution();
}
/// <summary>
/// this is the function you implement.
/// </summary>
/// <param name="sequenceA">the first sequence</param>
/// <param name="sequenceB">the second sequence, may have length not equal to the length of the first seq.</param>
/// <param name="resultTableSoFar">the table of alignment results that has been generated so far using pair-wise alignment</param>
/// <param name="rowInTable">this particular alignment problem will occupy a cell in this row the result table.</param>
/// <param name="columnInTable">this particular alignment will occupy a cell in this column of the result table.</param>
/// <returns>the alignment score for sequenceA and sequenceB. The calling function places the result in entry rowInTable,columnInTable
/// of the ResultTable</returns>
public int Align(GeneSequence sequenceA, GeneSequence sequenceB, ResultTable resultTableSoFar, int rowInTable, int columnInTable)
{
// Only fill in above the diagonal
if (rowInTable <= columnInTable)
{
return(0);
}
string a = sequenceA.Sequence;
string b = sequenceB.Sequence;
int m = Math.Min(a.Length, MaxCharactersToAlign);
int n = Math.Min(b.Length, MaxCharactersToAlign);
int[][] E = new int[2][];
E[0] = new int[MaxCharactersToAlign + 1];
E[1] = new int[MaxCharactersToAlign + 1];
// Initialize first row with cost of indels
for (int j = 0; j <= n; j++)
{
E[0][j] = 5 * j;
}
int previous = 0;
int active = 1;
for (int i = 1; i <= m; i++)
{
for (int j = 0; j <= n; j++)
{
var indels = E[previous][j] + 5;
if (j == 0)
{
// If first element, only have one option
E[active][j] = indels;
}
else
{
// Get the minimum cost from the two available indels and the diagonal match/sub
indels = Math.Min(E[active][j - 1] + 5, indels);
var diff = (a[i - 1] == b[j - 1]) ? -3 : 1;
E[active][j] = Math.Min(indels, E[previous][j - 1] + diff);
}
}
// Swap active and previous (new active will get overwritten)
active = (active == 0) ? 1 : 0;
previous = (previous == 0) ? 1 : 0;
}
// Return last element in last filled row
return(E[previous][n]);
}
public MainForm()
{
InitializeComponent();
m_dbController = new DatabaseController();
m_dbController.EstablishConnection("../../db1.mdb");
statusMessage.Text = "Loading Database...";
// Set the number of Sequences to load below.
m_sequences = m_dbController.ReadGeneSequences(10);
m_resultTable = new ResultTable(this.dataGridViewResults, m_sequences.Length);
statusMessage.Text = "Loaded Database.";
processor = new PairWiseAlign();
}
public MainForm()
{
InitializeComponent();
m_dbController = new DatabaseController();
m_dbController.EstablishConnection("../../db1.mdb");
statusMessage.Text = "Loading Database...";
// Set the number of Sequences to load below.
m_sequences = m_dbController.ReadGeneSequences(10);
m_resultTable = new ResultTable(this.dataGridViewResults, m_sequences.Length);
statusMessage.Text = "Loaded Database.";
processor = new PairWiseAlign();
}
public MainForm()
{
bool failed = false;
InitializeComponent();
statusMessage.Text = "Loading Database...";
// load database here
try
{
m_sequences = loadFile("../../" + GENOME_FILE);
}
catch (FileNotFoundException)
{
try // Failed, try one level down...
{
m_sequences = loadFile("../" + GENOME_FILE);
}
catch (FileNotFoundException)
{
try // Failed, try same folder
{
m_sequences = loadFile(GENOME_FILE);
}
catch
{
statusMessage.Text = "Failed to load database: " + GENOME_FILE + " not found.";
Refresh();
failed = true;
}
}
}
if (!failed)
{
string [] names = new string[NUMBER_OF_SEQUENCES];
for (int i = 0; i < NUMBER_OF_SEQUENCES; i++)
{
names[i] = m_sequences[i].Name;
}
m_resultTable = new ResultTable(this.dataGridViewResults, names);
statusMessage.Text = "Loaded Database.";
}
else
{
processButton.Enabled = false;
}
}
public MainForm()
{
bool failed = false;
InitializeComponent();
statusMessage.Text = "Loading Database...";
// load database here
try
{
m_sequences = loadFile("../../" + GENOME_FILE);
}
catch (FileNotFoundException)
{
try // Failed, try one level down...
{
m_sequences = loadFile("../" + GENOME_FILE);
}
catch (FileNotFoundException)
{
try // Failed, try same folder
{
m_sequences = loadFile(GENOME_FILE);
}
catch
{
statusMessage.Text = "Failed to load database: " + GENOME_FILE + " not found.";
Refresh();
failed = true;
}
}
}
if (!failed)
{
string [] names = new string[NUMBER_OF_SEQUENCES];
for (int i = 0; i < NUMBER_OF_SEQUENCES; i++)
{
names[i] = m_sequences[i].Name;
}
m_resultTable = new ResultTable(this.dataGridViewResults, names);
statusMessage.Text = "Loaded Database.";
}
else
{
processButton.Enabled = false;
}
}
//Dictionary<string, int> previouslyCalculatedValues = new Dictionary<string, int>();
/// <summary>
/// this is the function you implement.
/// </summary>
/// <param name="sequenceA">the first sequence</param>
/// <param name="sequenceB">the second sequence, may have length not equal to the length of the first seq.</param>
/// <param name="resultTableSoFar">the table of alignment results that has been generated so far using pair-wise alignment</param>
/// <param name="rowInTable">this particular alignment problem will occupy a cell in this row the result table.</param>
/// <param name="columnInTable">this particular alignment will occupy a cell in this column of the result table.</param>
/// <returns>the alignment score for sequenceA and sequenceB. The calling function places the result in entry rowInTable,columnInTable
/// of the ResultTable</returns>
public int Align(GeneSequence sequenceA, GeneSequence sequenceB, ResultTable resultTableSoFar, int rowInTable, int columnInTable)
{
// a place holder computation. You'll want to implement your code here.
//string key = sequenceA.Sequence.ToString() + sequenceB.Sequence.ToString();
//if (previouslyCalculatedValues.ContainsKey(key))
// return previouslyCalculatedValues[key];
// set up algorithm
initializeSequencing(sequenceA, sequenceB);
// calculate each additional row
//Overall time complexity of this part is O(n^2) and space complexity is O(n)
for (int i = 0; i < charA.Length; i++) // will go through the length of our frist sequence array size 0-5000 O(n)
{
resultRow = createNextRow(resultRow, charA[i], charB); //O(n)
}
// return score
//previouslyCalculatedValues.Add(key,resultRow[resultRow.Length - 1]);
return(resultRow[resultRow.Length - 1]);
}
public GeneNode Align(GeneSequence sequenceA, GeneSequence sequenceB, ResultTable resultTableSoFar, int rowInTable, int columnInTable)
{
// Check for equal alignment
if (rowInTable == columnInTable)
{
return(new GeneNode(null, '0', '0', 0));
}
// Linear space requirement
var resultSet1 = new List <GeneNode>(); // Prev Row
var resultSet2 = new List <GeneNode>(); // Current Row
// New Sequences
string seqA = "0" + sequenceA.Sequence;
string seqB = "0" + sequenceB.Sequence;
// Clean up the length to cap at 5000
int seqALength = seqA.Length;
int seqBLength = seqB.Length;
if (seqA.Length > MaxCharactersToAlign + 1)
{
seqALength = MaxCharactersToAlign + 1;
}
if (seqB.Length > MaxCharactersToAlign + 1)
{
seqBLength = MaxCharactersToAlign + 1;
}
// Core alignment algorithm
for (int i = 0; i < seqALength; i++)
{
for (int j = 0; j < seqBLength; j++)
{
GeneNode acc = null;
// Starting position
if (i == 0 && j == 0)
{
acc = new GeneNode(null, '0', '0', 0);
}
// Edge case
else if (i == 0 && j > 0)
{
// Get the node from the left cell, and add as indel
acc = new GeneNode(resultSet2[j - 1], '-', seqB[j], (resultSet2[j - 1].Cost + indel));
}
// Edge case
else if (i > 0 && j == 0)
{
// Get the node from the top cell, and add as indel
acc = new GeneNode(resultSet1[j], seqA[i], '-', (resultSet1[j].Cost + indel));
}
// Match Case
else if ((i > 0 && j > 0) && (seqA[i] == seqB[j]))
{
// Match or indel
var top = new GeneNode(resultSet1[j], seqA[i], '-', (resultSet1[j].Cost + indel));
var left = new GeneNode(resultSet2[j - 1], '-', seqB[j], (resultSet2[j - 1].Cost + indel));
var diag = new GeneNode(resultSet1[j - 1], seqA[i], seqB[j], (resultSet1[j - 1].Cost + match));
// Get node neighbor with smallest cost in order of: left, top, diag
acc = GetSmallest(left, top, diag);
}
// Subsitution Case
else if ((i > 0 && j > 0) && (seqA[i] != seqB[j]))
{
// Sub or indel
var top = new GeneNode(resultSet1[j], seqA[i], '-', (resultSet1[j].Cost + indel));
var left = new GeneNode(resultSet2[j - 1], '-', seqB[j], (resultSet2[j - 1].Cost + indel));
var diag = new GeneNode(resultSet1[j - 1], seqA[i], seqB[j], (resultSet1[j - 1].Cost + sub));
// Get node neighbor with smallest cost in order of: left, top, diag
acc = GetSmallest(left, top, diag);
}
// Add to lower row (current row) results
resultSet2.Add(acc);
}
// Make lower row the new upper row and clear the old lower row
resultSet1 = resultSet2;
resultSet2 = new List <GeneNode>();
}
// return the node at furthest (col, row) from origin
return(resultSet1[resultSet1.Count - 1]);
}
/// <summary>
/// this is the function you implement.
/// </summary>
/// <param name="sequenceA">the first sequence</param>
/// <param name="sequenceB">the second sequence, may have length not equal to the length of the first seq.</param>
/// <param name="resultTableSoFar">the table of alignment results that has been generated so far using pair-wise alignment</param>
/// <param name="rowInTable">this particular alignment problem will occupy a cell in this row the result table.</param>
/// <param name="columnInTable">this particular alignment will occupy a cell in this column of the result table.</param>
/// <returns>the alignment score for sequenceA and sequenceB. The calling function places the result in entry rowInTable,columnInTable
/// of the ResultTable</returns>
public int Align(GeneSequence sequenceA, GeneSequence sequenceB, ResultTable resultTableSoFar, int rowInTable, int columnInTable)
{
// Check for equal alignment
if (rowInTable == columnInTable)
{
return(0);
}
// Check Cache
if (InCache(rowInTable, columnInTable))
{
return(cache[rowInTable][columnInTable]);
}
// Linear space requirement
resultSet1 = new List <int>(); // Prev Row
resultSet2 = new List <int>(); // Current Row
// New Sequences
string seqA = "0" + sequenceA.Sequence;
string seqB = "0" + sequenceB.Sequence;
// Clean up the length to cap at 5000
int seqALength = seqA.Length;
int seqBLength = seqB.Length;
if (seqA.Length > MaxCharactersToAlign + 1)
{
seqALength = MaxCharactersToAlign + 1;
}
if (seqB.Length > MaxCharactersToAlign + 1)
{
seqBLength = MaxCharactersToAlign + 1;
}
// Core iteration
for (int i = 0; i < seqALength; i++)
{
for (int j = 0; j < seqBLength; j++)
{
int cost = 0;
// First cell, no cost value
if (i == 0 && j == 0)
{
cost = 0;
}
// Get the cost from the left cell, and add as indel
else if (i == 0 && j > 0)
{
cost = resultSet2[j - 1] + indel;
}
// Get the cost from the top cell, and add as indel
else if (i > 0 && j == 0)
{
cost = resultSet1[j] + indel;
}
else if ((i > 0 && j > 0) && (seqA[i] == seqB[j]))
{
// Match or indel
int top = resultSet1[j] + indel;
int left = resultSet2[j - 1] + indel;
int diag = resultSet1[j - 1] + match;
// Get smallest costs in order of: top, left, diag
cost = GetSmallest(top, left, diag);
}
else if ((i > 0 && j > 0) && (seqA[i] != seqB[j]))
{
// Sub or indel
int top = resultSet1[j] + indel;
int left = resultSet2[j - 1] + indel;
int diag = resultSet1[j - 1] + sub;
// Get smallest costs in order of: top, left, diag
cost = GetSmallest(top, left, diag);
}
// Add to the lower row.. the result set.
resultSet2.Add(cost);
}
// Make the lower row the new upper row
resultSet1 = resultSet2;
// Clear the old lower row
resultSet2 = new List <int>();
}
// Get result
int result = resultSet1[resultSet1.Count - 1];
// Add to the cache
AddToCache(rowInTable, columnInTable, result);
AddToCache(columnInTable, rowInTable, result);
// Return the new calculated cost
return(result);
}
/// <summary>
/// this is the function you implement.
/// </summary>
/// <param name="sequenceA">the first sequence</param>
/// <param name="sequenceB">the second sequence, may have length not equal to the length of the first seq.</param>
/// <param name="resultTableSoFar">the table of alignment results that has been generated so far using pair-wise alignment</param>
/// <param name="rowInTable">this particular alignment problem will occupy a cell in this row the result table.</param>
/// <param name="columnInTable">this particular alignment will occupy a cell in this column of the result table.</param>
/// <returns>the alignment score for sequenceA and sequenceB. The calling function places the result in entry rowInTable,columnInTable
/// of the ResultTable</returns>
public int Align(GeneSequence sequenceA, GeneSequence sequenceB, ResultTable resultTableSoFar, int rowInTable, int columnInTable)
{
// a place holder computation. You'll want to implement your code here.
return (Math.Abs(sequenceA.Sequence.Length - sequenceB.Sequence.Length));
}
/// <summary>
/// this is the function you implement.
/// </summary>
/// <param name="sequenceA">the first sequence</param>
/// <param name="sequenceB">the second sequence, may have length not equal to the length of the first seq.</param>
/// <param name="resultTableSoFar">the table of alignment results that has been generated so far using pair-wise alignment</param>
/// <param name="rowInTable">this particular alignment problem will occupy a cell in this row the result table.</param>
/// <param name="columnInTable">this particular alignment will occupy a cell in this column of the result table.</param>
/// <returns>the alignment score for sequenceA and sequenceB. The calling function places the result in entry rowInTable,columnInTable
/// of the ResultTable</returns>
public int Align(GeneSequence sequenceA, GeneSequence sequenceB, ResultTable resultTableSoFar, int rowInTable, int columnInTable)
{
string a;
string b;
//limit the number of characters in each string to 5000....O(1)
if (sequenceA.Sequence.Length > 5000)
{
a = sequenceA.Sequence.Substring(0, 5000);
}
else
{
a = sequenceA.Sequence;
}
if (sequenceB.Sequence.Length > 5000)
{
b = sequenceB.Sequence.Substring(0, 5000);
}
else
{
b = sequenceB.Sequence;
}
int[] topArray = new int[a.Length + 1];
//top Array begins as the top row of our table and is hard coded....O(n)
for (int i = 0; i < topArray.Length; i++)
{
topArray[i] = i * 5;
}
//bottom Array is where calculated scores are put
int[] bottomArray = new int[a.Length + 1];
int rowCount = 1;
//outer loop pulls out one character at a time, represented by one row in the table
//O(n) * inside the loop
//So the total is O(n^2)
foreach (char letter in b)
{
//the first column is hard coded...O(1)
bottomArray[0] = rowCount * 5;
rowCount++;
//inner loop iterates through each column, calculates the score
//the inside calculations are constant, so O(n)
for (int i = 1; i < a.Length + 1; i++)
{
int diff; //whether or not the characters in the row/col match
if (letter == a[i - 1])
{
diff = -3;
}
else
{
diff = 1;
}
//the score algorithm based on dynammic progamming and Needleman-Wunsch
int score = scoreMin(diff + topArray[i - 1], 5 + topArray[i], 5 + bottomArray[i - 1]);
bottomArray[i] = score;
}
//the pointers to the arrays are switched so the next row can be calculated based on
//the row above it. Bottom Array will be written over in the next loop
//Using the two array method gives us O(n) space complexity
int[] tempArray = topArray;
topArray = bottomArray;
bottomArray = tempArray;
}
//At this point the last index in topArray has the alignment score
return(topArray[a.Length]);
}
/// <summary>
/// this is the function you implement.
/// </summary>
/// <param name="sequenceA">the first sequence</param>
/// <param name="sequenceB">the second sequence, may have length not equal to the length of the first seq.</param>
/// <param name="resultTableSoFar">the table of alignment results that has been generated so far using pair-wise alignment</param>
/// <param name="rowInTable">this particular alignment problem will occupy a cell in this row the result table.</param>
/// <param name="columnInTable">this particular alignment will occupy a cell in this column of the result table.</param>
/// <returns>the alignment score for sequenceA and sequenceB. The calling function places the result in entry rowInTable,columnInTable
/// of the ResultTable</returns>
public int Align(GeneSequence sequenceA, GeneSequence sequenceB, ResultTable resultTableSoFar, int rowInTable, int columnInTable)
{
// a place holder computation. You'll want to implement your code here.
return(Math.Abs(sequenceA.Sequence.Length - sequenceB.Sequence.Length));
}
