queation about running the R script in parallel mode

[Mahnaz Tehrani]

Hi all,

I am testing some analysis directory by submitting the R script in parallel mode because my dcd file is so big and not doable to run via RStudio. I picked “hivp.dcd” as test:

library(bio3d)

library(igraph)

‌

dcd <- read.dcd("hivp.dcd")

pdb <- read.pdb("hivp.pdb")

‌

ca.inds <- atom.select(pdb, elety = "CA")

xyz <- fit.xyz(fixed = pdb$xyz, mobile = trj,

fixed.inds = ca.inds$xyz,

mobile.inds = ca.inds$xyz)

‌

rd <- rmsd(xyz[1, ca.inds$xyz], xyz[, ca.inds$xyz])

plot(rd, typ = "l", ylab = "RMSD", xlab = "Frame No.")

points(lowess(rd), typ = "l", col = "red", lty = 2, lwd = 2)

‌

# Root Mean Squared Fluctuations (RMSF)

rf <- rmsf(xyz[, ca.inds$xyz])

plot(rf, ylab = "RMSF", xlab = "Residue Position", typ="l")

‌

pc <- pca.xyz(xyz[, ca.inds$xyz])

plot(pc, col = bwr.colors(nrow(xyz)))

‌

# Cluster in PC space

hc <- hclust(dist(pc$z[, 1:2]))

grps <- cutree(hc, k = 2)

plot(pc, col = grps)

‌

# Cross-Correlation Analysis

cij <- dccm(xyz[, ca.inds$xyz])

plot(cij)

‌

after a short time, the calculation has been stopped at this point.

Any help would be appreciated.

Cheers

Zahra

‌

library(bio3d)

library(igraph)

dcd <- read.dcd("hivp.dcd")

NATOM = 198

NFRAME= 117

ISTART= 0

last = 117

nstep = 117

nfile = 117

NSAVE = 1

NDEGF = 0

version 24

|======================================================================| 100%

pdb <- read.pdb("hivp.pdb")

ca.inds <- atom.select(pdb, elety = "CA")

xyz <- fit.xyz(fixed = pdb$xyz, mobile = trj,

+ fixed.inds = ca.inds$xyz,

+ mobile.inds = ca.inds$xyz)

[Xinqiu Yao]

Hi,

‌

Can you tell what bio3d version and OS environment you used? For example, the output of sessionInfo(). Also, it would be helpful if you can try another computer (e.g., a Linux) and see if the problem persists.

[Mahnaz Tehrani]

Dear Xinqiu,
Hi and thanks for the answer.I have used it on linux (metacentrum supercomputer).
Here you can find technical informatin:

Thanks in advanced
Cheers
Zahra

uname -v
#1 SMP Debian 4.9.110-3+deb9u4 (2018-08-21)

sessionInfo()
R version 3.5.1 (2018-07-02)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Debian GNU/Linux 9 (stretch)

Matrix products: default
BLAS: /usr/lib/atlas-base/atlas/libblas.so.3.0
LAPACK: /usr/lib/atlas-base/atlas/liblapack.so.3.0

locale:
[1] C

attached base packages:
[1] stats graphics grDevices utils datasets methods base

loaded via a namespace (and not attached):
[1] compiler_3.5.1

BIO3D versin: 2.3-4

here is the submit.sh file have used to run the R-script..

#PBS -N helloR
#PBS -l walltime=2:00:00
#PBS -l select=1:ncpus=1:mem=4gb:scratch_local=5gb

#clean scratch after the end
trap 'clean_scratch' TERM EXIT

#INPUT directory - with R script and other inputs
MYIN=/storage/brno6/home/zahratehrani7/data-analysis-R/test/test-2
#INPUT R script
MYR=md.R
#OUTPUT directory - where to store results
MYOUT=/storage/brno6/home/zahratehrani7/data-analysis-R/test/test-2/mdRout
mkdir -p $MYOUT

module add R-3.5.1-gcc
export R_LIBS="/storage/brno6/home/$LOGNAME/Rpackages"
#export PATH=/storage/brno6/home/zahratehrani7/bio3d:$PATH

go to scratch directory

cd $SCRATCHDIR || exit 2

copy all files from $MYIN

cp $MYIN/* .

R -q --vanilla < $MYR > $MYR.output
#copy everything to MYOUT
cp * $MYOUT/

[Xinqiu Yao]

Hi,

‌

Have you tested it on a single computer? If it is only a problem in the parallel environment, you might need to contact your administrator of the cluster regarding the issue.

[Mahnaz Tehrani]

Dear Xinqiu,

The problem came from compiled MPI and has been fixed..

Cheer

Zahra

[Mahnaz Tehrani]

the issue has been resolved