WCGrid News and Talking Points!!

[[Ion]]

Anybody sporting these Ebola WU's yet? Is it still in Beta? I seem to have very limited success getting Beta's...

Yeah I have 63 pages of them

 

[t_ski]

^^^ That's where they all went...

 

[manofthem]

^^^ That's where they all went...

I've had a few beta OETs on my main rig today, not sure about the others.

 

[[Ion]]

^^^ That's where they all went...

Gotta get work.....not my fault the WCG scheduler is giving me some

 

[Peter1986C]

I had some OET WUs.

 

[Norton]

I have 4 wu's completed with 5 pages of them waiting to run

 

[Arjai]

Well, just discovered why i wasn't getting any OET WU's. At some point I reconfigured my projects page, for whatever reason, and did not reset it to all projects. So, I should get some, either this evening or tomorrow, after I have run some and make room for them!

OOPS!

 

[Peter1986C]

It was not auto-added to our profiles because it has the same special requirements thing that CEP has.

 

[Arjai]

Four articles written about Outsmart Ebola Together!
12/3 and 12/8/14

Mashable

KBPS San Diego

CNN Money

International Business Times

Boinc pointed to an ABC NEWS site but, searching their site yielded no results. So, to make four, I included the article from San Diego.

 

[Arjai]

Help Defeat Cancer

15 Dec 2014

Dear World Community Grid volunteers,

Since our last Help Conquer Cancer (HCC) project update, we have continued to analyze the results that you generated. Here, we provide an update on that analysis work, and new research directions the project is taking.

Volunteers for the HCC project received raw protein crystallization images and processed each image into a set of over 12,000 numeric image features. These features were implemented by a combination of image-processing algorithms, and refined over several generations of image-processing research leading up to the launch of HCC.

This traditional method of building an image classifier involves two types of learning: the crystallographer or image-processing expert (human), who studies the image and designs features, and the classifier (computer model), that learns to predict image labels from the designed features. The image classifier itself never sees the pixels;

More recently, we have applied a powerful computer-vision/machine-learning technology that improves this process by closing the feedback loop between pixels, features and the classifier: deep convolutional neural networks (CNNs). These models learn their own features directly from the image pixels; thus, they could complement human-designed features.

Figure 3 (below) shows CrystalNet's crystal-detection performance across 10 image classes in the test set. CrystalNet produces an area under curve (AUC) 0.9894 for crystal class classification. At 5% false positive rate, our model can accurately detect 98% of the positive cases.

Recently, however, HWI crystallographers were able to compile and share with us a complete record of all crystallization-trial proteins produced by the North-Eastern Structural Genomics (NESG) group. This dataset represents approximately 25% of all proteins processed by HCC volunteers on World Community Grid.

With more complete protein/cocktail information, combined with more accurate image labels from improved deep neural-net image classifiers, we anticipate greater success mining our protein-crystallization database. Work is ongoing.

Love to see our work creating interesting new processes to help with solving the sheer numbers we have given them!!

Complete news article, including even more confusing terms, can be found, HERE!

 

[Arjai]

Outsmart Ebola Together

12/19/2014

...The Ebola project, which debuted on the grid the first week of December, completed in one week what it would have taken a PC with a single processor about 35 years to accomplish.

Time magazine article about OET and WCG. I have no time to read it right now. But, you can!! Click HERE.

 

[Arjai]

Release Notes for BOINC 7.4
1/8/2015

Changes in 7.4.36

• Attaching to World Community Grid
• Back-up projects (0 Resource Share)
• Better detection of notice UPDATES (reduces the number of system notifications)
• Suspending GPUs should not suspend Bitcoin Miners
• Increasing the maximum number of coprocessor devices to 64
• Updates to OpenSSL(1.0.1j) and LibCurl(7.39.0)

Yea, NOW I can BitCoin and Crunch!!

Oh, here's a link....

And the Download page

 

[Peter1986C]

GPU mining XBT has been infeasible for some time now. Even special equipment is often not efficient enough to no lose money on the process with current value.

 

[twilyth]

Very true. And bitcoin hardware makers tend to run their machines for a few months before releasing them to the general public. When you factor in the rate at which the difficulty grows each month, you're lucky to get a few months worth of processing out of them before you start running at a loss. And if you factor in the cost of the hardware, you're probably running at loss to begin with. Now with prices a fraction of what they were a year ago, it difficult if not impossible to make a profit - at least for an individual trying to do this.

 

[Peter1986C]

World Community Grid Update - January 2015

By: Keith Uplinger
World Community Grid

16 Jan 2015

Summary An update about technical activities at World Community Grid in January 2015.

A brief update on some of the upcoming technical activity at World Community Grid:

In the next few weeks we will experience a few outages to upgrade our environment. First, we will be doing an upgrade to the network infrastructure with some new equipment, which will cause the entire website to go down for a few hours. We will also be upgrading our database servers next weekend. With this extra server capacity we will be exploring how to get the stats updates to run quicker.

On the application side of things, we are working on onboarding a few new science applications as well as fixing a few issues with current projects. As many of our members know, we do not release information about upcoming projects, so unfortunately I can not give details about these.

For the Outsmart Ebola Project, we are working to add mid job checkpointing to the VINA application. During our last BETA test, we identified that the checkpointing was causing more than the usual number of invalids compared to checkpointing at the job level. The reason for doing this is the next set of work units from the researchers are what are flexible docking work units and they can run long. Average job runtime for flexible work units is closer to 6 hours, where rigid jobs were running closer to 15 minutes. Once this is completed and we get a successful beta test, we will be starting the Outsmart Ebola Together project with a more consistent flow of work units.

The Uncovering Genome Mysteries project has an issue where it writes to the disk very frequently. At this time we are alpha testing some changes that will prevent this from happening. The current version in alpha does have a known issue and is being investigated, as it was causing invalid errors. Once we have a suitable version of the application, we will promote it to beta testing to make sure the changes are valid.

The VINA application on Android has seen an increase in errors due to the latest version of Android OS. It is causing a PEI exception to be thrown. We are working with Berkeley to get this resolved and tested in our environment. Currently we have recompiled the science application to run with PEI. Berkeley is also working on releasing a version of the BOINC client that will work properly with Android Lollipop.

As always, thank you for participating in World Community Grid and contributing to the humanitarian research projects that have benefitted from your generosity!

Source


-

 

[Peter1986C]

Surprising new prospects help advance the fight against neuroblastoma

By: Dr. Akira Nakagawara, MD, PhD
CEO of the Saga Medical Center KOSEIKAN and President Emeritus, Chiba Cancer Center

20 Jan 2015

Summary Prof. Nakagawara, lead researcher of the Help Fight Childhood Cancer project, updates us on the project's status after a remarkable year that included the publication of a breakthrough paper in 'Cancer Medicine'. World Community Grid results have revealed even more potential treatment mechanisms than the team previously thought, and paved the way for a second phase of the project.

Last year saw many successes for the Help Fight Childhood Cancer (HFCC) project in particular and the effort to overcome neuroblastoma in general. HFCC was a joint effort by personnel at the Chiba Cancer Center Research Institute (which I led) and Chiba University, (led by Dr. Hoshino and Dr. Tamura). Its primary goal was to identify small chemical compounds which target the particular sites of the proteins important in regulating the tumor cell growth and aggressiveness. Then, if the compounds we identified had the ability to kill the neuroblastoma tumor cells both in vitro and in vivo, the second goal was to develop new drugs to treat the patients with aggressive neuroblastoma using the structural information of the chemical compounds we identified. Thanks to your help, we have made significant progress toward both of those goals. We greatly appreciate all of the World Community Grid volunteers for their enormous enthusiasm to help cure and support children with cancer.

Overview of our recent progress

As we reported last year, we had great success in identifying some promising anti-cancer compounds, and have published the first round of our results in the journal Cancer Medicine. However, these were only a subset of the promising results that we have found. So far, World Community Grid members have helped us identify anti-cancer potential in small chemical compounds that work in three different ways:

• 
  1.TrkB receptor antagonists:
  These were the subject of our Cancer Medicine paper. We used your AutoDock computations to identify 7 chemicals from a library of 3 million chemical compounds. Those chemicals showed very low IC50 values (low values indicate a measurement of more effectiveness) killing neuroblastoma cells in vitro. We then selected the 2 chemicals with the lowest IC50 values, and found that they significantly suppressed the zenografted tumor growth in nude mice. Dr. Hoshino's group at Chiba University is currently generating new small chemical compounds based on the structure of the candidate compounds you helped us identify. These are chemical compounds with much lower IC50 value to kill the neuroblastoma cell, and are based on the structure of those identified by World Community Grid.
  
  The primary short-term challenge for this aspect of our research is finding a pharmaceutical company that can collaborate with us in turning these promising compounds into a medication. This has been difficult because although neuroblastoma is terrible, in absolute terms it is still a very small market, and few companies are interested in devoting development resources to it.
  
  2. TrkB receptor agonists:
  As a byproduct of World Community Grid computations, we happened to find the agonists of TrkB which appear to function similarly to Brain-derived neurotrophic factor (BDNF), which is a physiological ligand of TrkB. This could provide another avenue of attack against neuroblastoma. The functional analysis of these results is going on now in my laboratory.
  
  3. Antagonists inhibiting the binding sites of ALK receptor and ShcC adaptor protein:
  We have previously found that both the ALK receptor and its adaptor protein ShcC are markers for aggressiveness in neuroblastoma. The ALK protein interacts with the ShcC protein at two binding sites. Therefore, we were looking to block those sites by identifying small chemical compounds which interrupt the binding sites. Calculations done on World Community Grid successfully identified several compounds for the two binding sites which kill neuroblastoma cells in vitro with low IC50 values. We are currently analyzing the molecular pathway by which those compounds kill the tumor cells.
  
  For these compounds, we collaborated with Dr. Sakai's group at the National Cancer Center Research Institute in Tokyo. The student who was working on this project has left the lab, so we are now looking for the new researcher who is interested in this project.

Sharing our research

The process of making this data public is currently ongoing - the TrkB antagonists were the subject of our paper in Cancer Medicine in 2014, and that raw data is publicly available. We are still working on the other data on TrkB agonists and ALK/ShcC antagonists. We presented our Phase 1 results at several conferences in Japan, as well as at the ANR (Advances in Neuroblastoma Research) meeting held in Germany in 2014. That raw data is not yet publicly available.

Next steps

As I mentioned last July, I moved from Chiba to Saga and took up a position at the Saga Medical Center KOSEIKAN, whose hospital was founded in 1834 during the Edo Days of Japan. Therefore, the HFCC team is now a collaboration between the Graduate School of Medicine, Chiba University, and the Saga Medical Center KOSEIKAN. The home page of HFCC will soon be moved from the Chiba Cancer Center to the Saga Medical Center KOSEIKAN.

Phase 1 of HFCC targeted neuroblastoma, and we are now developing a Phase 2 that will target many more childhood cancers. This future work will include a new team from Hong Kong University led by Dr. Godfrey C.F. Chan, a pediatric oncologist. We hope to bring the next phase of this project to World Community Grid soon.

Once again, thank you to the whole World Community Grid team, and to all the volunteers, for making this research possible. With your help, we are much closer to finding effective treatments for a devastating childhood disease.

---------------------

Source: http://www.worldcommunitygrid.org/about_us/viewNewsArticle.do?articleId=413

---------------------

 

[twilyth]

The BDNF connection is interesting since it has been implicated in depression and IIRC there is some research that implies that the way serotonin uptake inhibitors really work is via BDNF. Reductions in BDNF seem to result in atrophy of the hippocampus which facilitates the creation of memories. Persistent stress seems to reduce BDNF levels which results in depression as per the HPA axis theory.

 

[Peter1986C]

The end of the beginning is near for FightAIDS@Home

By: The FightAIDS@Home research team

20 Feb 2015

Summary Thanks to the incredible generosity of World Community Grid volunteers, the FightAIDS@Home project team has finished with an important stage of their project. The research team has refocused on analyzing their existing results and preparing for the end of this historic grid computing stage.

FightAIDS@Home has been running on World Community Grid in some form almost since the beginning of World Community Grid itself: our project launched in 2005. Thanks to the enormous and ongoing support of our worldwide community of volunteers, we have expanded the scope of our research and explored new targets and drug candidates that we simply could not imagine at the outset. It hardly seems sufficient to say thank you for donating over 330,000 years of processing time to support our research, but once again, from all of us to all of you: thank you. Clearly we could not do this research without you.

With your help, we have reached a new milestone: no new AutoDock (AD) or AD Vina docking experiments are currently being generated. Put another way: we’re done creating new work tasks. The AutoDock queue is now empty, and the AD Vina queue has more than a year's worth of jobs left. Most of our efforts have shifted towards analysis.

The analysis of the FightAIDS@Home data has several levels of difficulty due to the sheer amounts of data, which are comprised of several structures of any drug target as well as millions of small molecules, resulting in hundreds of millions of data points. We are attempting to use a couple of approaches to mine this data, one of which includes examining amino-acids involved in top-ranked dockings. Another approach is to investigate the atomic coordinates of important interactions (pharmacophore) between the protein and the small molecule that was docked. Figures 1 and 2 (below) illustrate a simple example of inhibitor TL3 (Figure 1) and the predictions of 1 experiment (Figure 2, ~5.5 million dockings on 1 protein structure). Of course, these evaluations must be done with a large set of known inhibitors and across myriad protein structures. Once these methods pass a high level of confidence, molecules will be bought and sent to collaborators to be tested.


Figure 1. Docked pose of known HIV-1 protease inhibitor TL3 in an HIV-1 protease structure (not shown). Spherical representations (accompanied with dots, orange for TL3, green for a water molecule) represent important locations for protein-ligand interactions that are used to evaluate if a molecule may be a good drug candidate. The green sphere represents the location of an important ("flap") water molecule often observed in HIV-1 protease co-crystal structures. The 2 orange spheres directly below the green sphere represent two locations of an interaction with significant amino acids (Asp25) of HIV-1 protease.




Figure 2. Same docked pose of TL3 in HIV-1 protease as Figure 1 with top percentage of interactions from 1 experiment (pink spheres) and several predictions (transparent surfaces) for important protein-ligand interactions. Note that the water molecule (green) and the 2 orange interactions below it are always predicted.

http://www.worldcommunitygrid.org/about_us/viewNewsArticle.do?articleId=420Link

 

[Arjai]

Using one cancer to help defeat many: Mapping Cancer Markers makes progress
12 Feb 2015

Results from the first stage of the Mapping Cancer Markers project are helping the researchers identify the markers for lung cancer, as well as improve their research methodology as they move on to analyze other cancers.

Once again, the Mapping Cancer Markers (MCM) team would like to extend a huge thank you to the World Community Grid members. Although we publish this thank you each update, we are truly grateful for your contribution to this project.

New stage in lung cancer signature discovery

The MCM project has continued to process lung cancer data, exploring random fixed-length signatures of between 5 and 25 biomarkers. This computational component of the “landscape” stage is winding down, and we are preparing to transition our focus to a narrower set of genes of interest. Target genes will be selected by integrating results from several methods, carefully combining statistics from the initial results with pathway and biological-network analysis.

Figure 1: An iterative strategy for biomarker discovery. Work units are processed on World Community Grid. The results are analyzed via a Streams pipeline. This generates a list of high-scoring genes, which combined with biological network information (NAViGaTOR) are used to design new MCM work units targeting areas of interest in signature space.

Full Story

 

[Arjai]

World Community Grid Update - February 2015
20 Feb 2015

In the past month, you may have noticed a few things that have been completed or updated with the World Community Grid website and servers. We were able to release the Outsmart Ebola Together (OET) application withupdated checkpointing code. This has resulted in an increased flow of work units to our members. At the current pace, we have plenty of work to last us for awhile. We have also released the updated Android application that runs on 4.1+ only. This was released for both the OET and FightAIDS@Home(FAAH) projects.

We have also successfully completed two hosting environment changes. The first one was for our network infrastructure, and the second one was an update to our database servers. Members have already noticed the increase in speed for daily stats updates, as a result of phase 1 of this upgrade.

More Techy stuff!!

 

[manofthem]

Great posts @Arjai!

Just read (skimmed really) through those articles, and those are exciting updates.

I have noticed an increase in OET wus lately. Also, I didn't know that UGM did a lot of writes to disk, but apparently they're working on a solution to fix that. Pretty cool stuff.

 

[Arjai]

Making progress against two of the world’s deadliest diseases
9 Feb 2015

SummaryThe GO Fight Against Malaria team has just published a paper in which they describe an unexpected benefit of their research: they’ve identified two compounds that could advance the future development of new drugs to treat tuberculosis, including drug-resistant tuberculosis. These results open up a new front in the fight against tuberculosis, which is constantly evolving to resist existing treatments.

New fragment-sized inhibitors of the TB drug target called "InhA" were discovered in GO FAM experiment 5. The cyan ball-and-stick molecule in the center shows the predicted binding mode of the most potent inhibitor we discovered, while the InhA enzyme it inhibits (which was the target of the docking calculations) is shown in grey.

Paper Title:

"A Virtual Screen Discovers Novel, Fragment-Sized Inhibitors of Mycobacterium tuberculosis InhA"

Lay Person Abstract:

As part of the GO Fight against Malaria (GO FAM) project, some of the calculations performed using the power of World Community Grid were applicable to Tuberculosis. These calculations identified several chemical compounds which have the potential to facilitate the future development of new types of compounds to help with the fight against Tuberculosis, a bacterial disease that kills 1.5 million people each year and is becoming more resistant to established drug treatments. Further laboratory testing shows that two of these compounds could possibly pave the way to the development of new drug treatments that would not be disabled by a prevalent form of drug resistance that has emerged in tuberculosis. This is good news considering that Tuberculosis is on the rise and kills more people than any other bacterial disease.

Nice!!

More Story here

 

[Arjai]

Computing for Clean Water on the road to publication
9 Feb 2015

SummaryThe Computing for Clean Water team has written a paper describing the results of their research on World Community Grid. They’ve described the novel flow effect that all you volunteers helped discover. Their paper is currently under consideration at a prestigious journal.

The scientific process has many phases, and one of the most challenging of these is publishing results. What often amounts to years of a scientist’s working life has to be distilled to just a few pages, and done so in a way that is both clear and compelling - at least for other experts in the field. The team behind Computing for Clean Water worked for many months last year on drafting and polishing such an article, which sums up several years of work. Indeed, if we included all the processing time that you, the volunteers, have contributed to the project with your PCs and laptops, we could argue that the article represents many thousands of years of collective effort!

We have been waiting several months now to see whether we are amongst the lucky few, or whether we will have to revise our article and perhaps consider submitting it to a less demanding journal. Ultimately, the goal is to get the information out there so other people can benefit from it. So there is always a balance between wanting to ensure the broadest possible audience for our results by publishing in a top journal, and simply ensuring that the information is accessible to other scientists, by publishing in a more lenient one. In the case of Computing for Clean Water, it’s fair to say that the whole team behind this project feels an additional responsibility to the large community of volunteers, to do the best possible job of promoting their diligent efforts.

So thanks again to all the volunteers on Computing for Clean Water for your help and for your patience with this process. Rest assured that you will be the first to know when and where the results will be published.

Best of Luck, to them!!

Rest of the story...

 

[Arjai]

Database Upgrade: Saturday, February 28, 2015 at 03:00:00 UTC
24 Feb 2015

Word for Word:

Summary
Database upgrades will be performed Saturday, February 28th.

World Community Grid will be performing a database upgrade starting Saturday, February 28, 2015 at 03:00:00 UTC. The window for this maintenance activity is estimated to be 16 hours, although we anticipate the actual outage time will be less.

During this database upgrade, volunteer devices may not be able to fetch new research tasks or return completed work for a period of time. However the World Community Grid website will be available.

No action is required by the members, as the BOINC/World Community Grid software application will automatically reconnect to our servers once the upgrade is over.

Thank you again for your participation in World Community Grid!

So, Yea. Let's get those buffers pumped up an extra day. AND Be sure to Manually update Friday Night, or Day, if'n you prefer that!

Proof.

 

[Arjai]

Seven quadrillion comparisons later, Uncovering Genome Mysteries is just getting started
26 Feb 2015

...The project expects to examine more than 200 million proteins, the majority of which were generated in environmental and ecological studies ranging from bacteria in marine ecosystems in Australia, to Amazon River samples from Brazil. Similarity data from these comparisons will lead to a better understanding of metabolic and structural functions of the predicted proteins in databases, and uncover many new features and cellular processes in microorganisms. Of the expected 20 quadrillion (20,000,000,000,000,000) comparisons in the project, about 36% have been completed thus far, equivalent to almost 8,000 CPU-years of computation.

Volunteers participating in the UGM project process work units that contain sets of protein sequences predicted from a variety of organisms, and compare those against each other. Every time a significant similarity between two sequences is detected, a line of output is written that contains the coordinates and information on the statistical significance of the similarity. All of the output data together allow us to trace functional predictions of unknown sequences when they are similar to sequences with known functions, and indicate how organisms and their biochemistry, metabolic functions, and other cellular processes relate to one another.

...Many applications in health, environment, and agriculture can be attributed to making use of such data. For example, they enabled the development of new strategies to fight pathogens that threaten human and animal health, and development of diagnostics, treatments, and preventions through appropriate design of vaccines. But there are many other applications to be discovered, in agriculture, industry or the environment, through the study of the wide variety of proteins and enzymes.

An awesome undertaking, already making waves!!

Even more good news, here!!