Infinite Mixture ModelBased Clustering of DNA Microarray Data Using openMP

1
Infinite Mixture Model-Based Clustering of DNA
Microarray Data Using openMP
2
DNA microarrays
3
Clustering MA data (not computers)
Conditions
Clustering
Genes
4
Why do I need sooo much computing power?

• Goal Determine all posterior pairwise
  probabilities of two genes/samples belonging to
  same cluster
• IMMs cannot be solved analytically
• Use sampling method to approximate posterior
  probabilities
• Typically
• 10,000 iterations
• Sevaral 1,000 genes
• 10 200 samples
• Compute O(genes2 x samples2) probabilities per
  iteration
• Some overhead for cluster-reassignment, other
  model parameters

5
Open Multi-Processing (OpenMP)

• Facilitates parallelization of C and Fortran
  code for shared memory environments
• e.g. multi-processor machines
• Set of compiler directives, system variables, and
  library functions
• Platform-independent
• Website http//www.openmp.org/
• Parallelize sequential code by using compiler
  directives
• Relatively small programming effort
• Reduced risk for programming errors
• Use of shared memory
• Reduces the communication overhead required to
  synchronize multiple threads
• But cannot run threads on multiple nodes

6
Hello World in openMP

• include ltstdio.hgt
• include ltomp.hgt
• int main(int argc, char argv)
• int id, nthreads
• pragma omp parallel private(id)
• id omp_get_thread_num()
• printf("Hello World from thread d\n",
  id)
• pragma omp barrier
• if ( id 0 )
• nthreads omp_get_num_threads()
• printf("There are d
  threads\n",nthreads)
• return 0

7
OpenMP examples

• Header
• include ltomp.hgt
• Library Functions
• omp_set_num_threads(4)
• printf("number of threads d\n",
  omp_get_num_threads())
• Compiler directives
• pragma omp for schedule(dynamic, 1)
• for(j0jltQj)
• clusterProbabilitiesj getProbCsMissing2(i,j
  ,Contexts)

8
OpenMP examples

• More Compiler directives
• pragma omp for
• for(j0jltQj)
• pragma omp critical
• sigmasij1.0/gengam(betai vi/2.0,betai
  /2.0)

9
OpenMP examples

• More Compiler directives
• int i
• pragma omp parallel for private(i, pos)
• for(j0iltTi)
• same as
• pragma omp parallel for private(pos)
• for(j0iltTi)
• int i

10
Some more compiler directives

• Reduction
• pragma omp do reduction (sum)
• ? summarize the share variable sum
• Parallel region
• pragma omp parallel
• Sections
• pragma omp sections
• pragma omp section
• Code block 1
• pragma omp section
• Code block 2

11
The making-of

• Start an interactive session
• jfreuden_at_fructosegt qsub -I -l nodes1opteron
• Intel compiler
• jfreuden_at_bmi-opt2-01gt module load openmpi-intel
• jfreuden_at_bmi-opt2-01gt icpc -w openmp
• g compiler
• jfreuden_at_bmi-opt2-01gt module load gcc-4.2.3
• jfreuden_at_bmi-opt2-01gt g -fopenmp

12
The batch file

• PBS -S /bin/csh
• PBS -l nodes1opteronppn2
• PBS -l walltime180000
• PBS e /users/jfreuden/runGimm/stderr.txt
• PBS -o /users/jfreuden/runGimm/stdout.txt
• setenv OMP_NUM_THREADS cat PBS_NODEFILE
  grep HOST wc -l
• module load intel
• cd /users/jfreuden/runGimm/
• R CMD BATCH runGimm.R

13
Simulation study Non-informative samples

• 4 gene clusters of sizes 20, 20, 80, and 80
• 3 sample clusters of size 5
• Additional samples
• m 5, 10, 20, 50, 100
• No change in expression
• Same noise level
• 100 repeats for each level

14
Simulation study Non-informative samples
15
Simulation study Non-informative samples
16
Simulation study Non-informative samples
17
Questions?Comments?
18
Additional Slides
19
Clustering
Dhaeseleer (2005)
20
Example for Gibbs Sampling BUGS
21
Simulation study Simple Case
22
Simulation study Simple Case
23
Simulation study Time course 1
24
Simulation study Time course 2
25
Simulation study Non-informative samples
26
Simulation study Non-informative samples