Posted 19 Feb 2018 by
FrancescoAsnicar [SSC11]
Hi everyone,
First of all, thank you very much for your continued support to our project, we really appreciate all the contrbutions.
We started a new set of experiments about the Pseudomonas aeruginosa (strain PAO1) with the aim of producing the expansion of each of the single gene present in this organism.
The final goal is to provide these results to biologists for studying new possible interactions.
Many thanks,
Francesco
Posted 29 Jun 2017 by
FrancescoAsnicar [SSC11]
We are still waiting for some validations from the biologist involved in this project, based on the bunch of workunits that we ran.
This step is fundamental to understand if we need to adjust some parameters before starting doing the complete set of experiments.
I hope we can start soon with these experiments.
Many thanks to you all for your support to our project,
Francesco
Posted 17 Apr 2017 by
FrancescoAsnicar [SSC11]
Any roadmap? I'm continuing to download vitis vinifera....
We just submitted to system a bunch of experiments with very different parameter values. We'll now collect and process them. Then, we need to evaluate the results to find the best set of params to use for the "real" experiments. Some of the experiments were too fast to compute and there were users complaining about the bandwidth consumption, so we need to carefully define these new experiments to be suitable within the BOINC environment.
I believe that we can start again new experiments on Pseudomonas aeruginosa by the end of May.
Posted 10 Apr 2017 by
FrancescoAsnicar [SSC11]
Hello all,
here a brief description of some new experiments coming out in the next days!
Pseudomonas aeruginosa is a bacterium that can be harmful for plants and animals, including us, humans. P. aeruginosa is in general a multidrug resistant bacterium that have been associated with serious illnesses. (https://en.wikipedia.org/wiki/Pseudomonas_aeruginosa)
For this reason, the World Health Organization (WHO), early this year, included P. aeruginosa in its list of bacteria for which there is urgent need of antibiotics. (http://www.who.int/mediacentre/news/releases/2017/bacteria-antibiotics-needed/en/)
Thanks to a collaboration we are starting with a biologist, who is specifically studying this bacterium, we will start in the next days a bunch of new experiments. The first experiments will test several different parameters using a very well known (by the biologists) gene network of P. aeruginosa that controls the number of bacterial cells, named quorum sensing (https://en.wikipedia.org/wiki/Quorum_sensing). If everything turns out well, with our results we can help the biologist we are collaborating with, as she will study our predictions in her in-vitro experiments! :)
Many thanks again to all of you for your support to our project.
Posted 6 Mar 2017 by
FrancescoAsnicar [SSC11]
Hi all,
I'm trying to compile the new optimized app for the Mac platform, but I'm having some trouble and I hope that there are someone that can help me out.
I set up the BOINC repository and compile all the external dependencies (c-ares, libtool, openssl, etc.). Then, I compile BOINC (library only) and then I was trying to fix the Makefile_mac present in the repository (https://bitbucket.org/francesco-asnicar/pc-boinc). However when it comes to compile the application I got the following error:
pc.cpp:119:6: error: unused function 'getCorrelationCoefficient4' [-Werror,-Wunused-function]
void getCorrelationCoefficient4(double a1, double b1, double c1, double a2, double b2, double c2,
^
Does anybody have any idea how to solve this?
Many thanks,
Francesco
Posted 1 Feb 2017 by
FrancescoAsnicar [SSC11]
Hi all,
first of all I want to thank you all for your computational support to our project, this is fantastic!!
I just want to provide you some info on the new "oge_dream5_net3" experiments that are coming out. These are experiments on the E. coli organism, but with a different (smaller) dataset w.r.t. the other experiments on E. coli.
This new smaller dataset is one of the dataset used in the DREAM5 competition, which means that with these experiments we can fairly compare our Gene Network Expansion approaches against other competitors around the world.
If you want more info, I'm more than happy to discuss with you about these experiments.
Many thanks,
Francesco
Posted 11 Jan 2017 by
FrancescoAsnicar [SSC11]
Hi Daniel,
about this:
I found something what may be a problem. You use undirected graph, so I thought that I could reduce number of iterations of loop at pc.cpp:418 to test (i,j) pairs for j > i only. However after doing this output file size changed from 47.8K to 67.6K. Original code before my changes also generated bigger file after applying this change. I checked code briefly and do not see anything obvious what may cause this. Could you take a look on this?
it is actually not possible to half the number of iteration because the algorithm choose a pair of node i,j and test whether the arc linking the two nodes should be removed. When l increases, the test is conditioned to a set of neighbours of size l of the first node. If the edge is not removed, it could be the case that there exists a set of neighbours of j of a certain size l that allows the removal of the edge. So, it is important to test all the possible combination of i,j.
I hope it is clear enough. A bit more details can be found here: http://www.jmlr.org/papers/volume8/kalisch07a/kalisch07a.pdf (page 5, Algorithm 1)
Many thanks,
Francesco
Posted 19 Oct 2016 by
FrancescoAsnicar [SSC11]
Hi all,
I just wanted to link here the public repository of the source code of the BOINC-client implementation that is running on your machines!
https://bitbucket.org/francesco-asnicar/pc-boinc/
We are open to improvements or comments on the code.
Many thanks to you all for supporting our project!
Posted 24 Aug 2016 by
FrancescoAsnicar [SSC11]
Hi all,
good news from the scientific world!
We just received the notification that another work, whose results were made possible thanks to the gene@home project, has been published!
For the one fo you interested in this new publication, here the reference:
NES2RA: Network expansion by stratified variable subsetting and ranking aggregation
published in the
International Journal of High Performance Computing Applications (doi:10.1177/1094342016662508, http://hpc.sagepub.com/content/early/2016/08/17/1094342016662508.full.pdf+html)
Many thanks to all of you,
Francesco
Posted 6 May 2016 by
FrancescoAsnicar [SSC11]
Hi all,
the current running experiments on the gene@home are part of the OneGenE project.
OneGenE aims at expanding each single gene of an organism.
The main idea of OneGenE is that if we compute the gene expansion list for each gene of an organism, we are then able to use these expansions multiple times, to produce the final expansion list of any local gene network.
With the previous gene network expansion methods, the results that we computed are affected by the knowledge of the local gene network inserted in each experiment. This means that, when we want to expand a different local gene network we cannot use the results that we already computed on the same organism. We need to schedule new ad hoc experiments for the new local gene network of interest.
With OneGenE, we should be able to build a public database containing the single expansions for each gene of an organism, possibly in the future, for more than one organism.
With this public database, we could offer the possibility to produce a gene network expansion list, by combining the single expansion already computed.
The first organism that we selected for the OneGenE project, that is running on the gene@home project, is Escherichia coli.
As of today, we computed about 600 single gene expansions out of about 4100 genes present in the E. coli, thanks to your support to the gene@home project.
If you have any doubts or you want more information, we are available to discuss.
Many thanks,
Francesco
Posted 27 May 2014 by
FrancescoAsnicar [SSC11]
Hi all gene@home supporters!
We just empty our experiments queue, do not worry new works will be available soon! ;-)
For the moment we are collecting and processing all the results (just few are not yet come back). We are also checking the feedback provided for the application, in order to see if there are things that should be fixed or improved.
We hope in a few days to put here http://gene.disi.unitn.it/test/forum_forum.php?id=2 some results from this last experiments.
Thank you all for crunching! :)
Posted 23 May 2014 by
FrancescoAsnicar [SSC11]
Thanks for reporting the issues. We are collecting all issues reports, and at the end of this second test phase we investigate each one in order to fix the application.
Can I ask you: it was just that work-units that report the problem or you get other work-units with the same error?
Thank you very much
Posted 22 May 2014 by
FrancescoAsnicar [SSC11]
In order to validate our algorithm, we started several experiments, friendly called "empty" (because they didn't have a local gene network that will be expanded).
These experiments should tell us the impact of the "divide-et-impera" approach performing inference, with respect to the expansion done within every subgraphs.
Posted 14 May 2014 by
FrancescoAsnicar [SSC11]
To the initial work programmed for this second phase of parameters tuning, we added a quite big deal of experiments.
In fact, at the end of the last week we received a new local gene network related to flavonoids. This was a good news because we can expand this new local gene network using the same input used for the first local gene network (FOS), since both of them are related to the flowering of the Arabidopsis thaliana.
We posted a gif of the most significant plots we obtained from the first phase of parameters tuning here: http://gene.disi.unitn.it/test/forum_thread.php?id=85
Enjoy!
Posted 13 May 2014 by
FrancescoAsnicar [SSC11]
The work-unit id is 1524 not 967, right?
Here the link: http://gene.disi.unitn.it/test/workunit.php?wuid=1524
We are investigating the fact, and it is quite strange since both pc are Windows x64!
Thanks for posting :)
Posted 15 Apr 2014 by
FrancescoAsnicar [SSC11]
We update the work generator, now ti should be able to better estimate the computation time needed for each work unit. We fix also a minor bug during the write (hence changing the value read) of the score value in the checkpoint file.
Posted 13 Apr 2014 by
FrancescoAsnicar [SSC11]
I report the fixes and changes in the "change log" thread under "development" section of the forum.
Posted 26 Mar 2014 by
FrancescoAsnicar [SSC11]
Ok, thanks for reporting the issues.
We find that we forgot to specify a <no_delete /> tag in the input template.
Now we have updated the work generator, we stop the file deleter daemon and we copy back the experiments files in the download directory.
This should allow us to finish the current tests. We will try in a few days the updated work generator.
Posted 25 Mar 2014 by
FrancescoAsnicar [SSC11]
Hi there,
we should have solved the issues arise yesterday related to the not unique filename generated by the work generator.
We also update the application in order to get a bette stderr from the executions.
Let us know if there are still problems,
Thank you all for the support!
Posted 25 Jan 2014 by
FrancescoAsnicar [SSC11]
As I wrote on the BOINC @ TN thread on BOINC.Italy the 32 bit Linux executable was compiled on Ubuntu with kernel 3.2.0-56 and glib version 2.13, while the 64 bit was compiled on Ubuntu with kernel 3.2.0-23 and glib version 2.13..
I can't figure out why the 32 bit is resolved while the 64 bit still present the known issue related to the glib version.
Hope we can find out soon and solve this problem.
Thanks again for report here all the problems you guys found!