Simon Fraser University (SFU) - CZSaw, IMAS & Tableau

VAST 2010 Challenge
Grand Challenge: Arms Dealing and Pandemic

Authors and Affiliations:


Dustin Dunsmuir, Simon Fraser University,
Saba Alimadadi, Simon Fraser University,
Mahshid Z. Baraghoush, Simon Fraser University,
Victor Chen, Simon Fraser University,
Minoo Erfani Joorabchi, Simon Fraser University,
Mona Erfani Joorabchi, Simon Fraser University,
Eric Lee, Simon Fraser University,
Cheryl Qian, Purdue University,

John Dill, Simon Fraser University,
Chris Shaw, Simon Fraser University,
Robert Woodbury, Simon Fraser University,


CZSaw is a visual analytics tool for sense-making across text documents with extracted entities that focuses on the analysis process and provides a variety of flexible data visualizations. It records the analysis process and model, and visualizes them in a history view (providing quick access to past states in the process) and dependency graph respectively. The dependency graph allows quick rerunning of parts of the analysis process on new data. What makes this possible is capturing all semantically meaningful interactions in a script language which can be edited by an expert.
CZSaw was developed in the School of Interactive Arts and Technology at Simon Fraser University by Victor Chen, Dustin Dunsmuir, Eric Lee, Nazanin Kadivar, Cheryl Qian, John Dill, Chris D. Shaw, and Rob Woodbury. A paper entitled “Capturing and Supporting the Analysis Process” was presented at VAST 2009; the presentation materials can be found on the vgtc website. For more information, see CZSaw's webpage.

IMAS (The Interactive Multi-genomic Analysis System) is a visual analytics system for knowledge discovery in genomic information, which provides the user with a horizontal zooming of the sequences. Initially it was developed by Chris Shaw and his students in 2007. IMAS visualizes the output of some common bioinformatics tools such as BLAST and ClustalW in a unified framework. BLAST is used for pair-wise alignment, and ClustalW is used as our multi-alignment program.

Version 5.1 of Tableau was used for solving Mini Challenge 2. Tableau data visualization software was a project in Stanford University in 1997 by Chris Stolte and Professor Pat Hanrahan. Tableau Desktop was released in 2003. With Tableau, analysts can connect to large amounts of data and auto-generate visualizations including reports, tables, charts and graphs through a drag-and-drop interface.



Video for grand challenge



GC: Investigate any possible linkage between the illegal arms dealing and the pandemic outbreak. 

In particular, you need to address the following:

1. Briefly describe your hypothesized linkage between the arms dealing activity and the pandemic outbreak. 

2. Where did the disease originate (as far as you can tell with the data)?  Provide a time-based tracing of its spread among countries.  If you have established linkages between arms dealers in certain countries and the pandemic, please indicate this. 

3. We had countries with arms dealers identified in MC 1 that did not suffer pandemic outbreaks in MC 2.  Provide a hypothesis as to why some countries that may have been involved with arms dealers did not suffer an outbreak? 

In the first half of 2009, there was a major epidemic outbreak (the Drafa virus) that spanned over several countries around the world. Through careful investigation we have developed strong hypotheses about its origin and how it spread to many different countries.

The country of origin for the outbreak of the disease was Nigeria since this is the native sequence which we determined is the ancestor of all the outbreak sequences for the countries of western Africa. The disease first spread in many countries in western Africa before jumping to countries in eastern Africa, Asia, and South America in April and May 2009. It is our hypothesis that this jump to more developed countries occurred as the result of arms dealers contracting the disease from someone from Nigeria. They then unintentionally spread it back to their home countries.

Much intercepted communication from the year leading up to April 2009 shows that a group of arms dealers agreed to meet many of their customers in Dubai (UAE) within the same week (from April 15th 2009 to April 22nd 2009). There is no indication that many of these customers knew of each other, so this was most likely a series of meetings rather than a larger conference involving everyone. This is supported by the fact that many customers planned to arrive and leave on different days throughout the week.

From multiple phone conversations, surveillance reports and emails it is clear that the core group of arms dealers were Nicolai Kuryakin (likely from Russia), Mikhail Dombrovski of Russia, Arkadi Borodinski of Ukraine, and Boonmee Khemkhaeng of Thailand. These people planned to attend almost all the meetings during the week. Hence, we can infer that there were some physical connection paths between all the people who attended at least one of the meetings in Dubai. One of the people that came for a meeting was Dr. George Ngoki from Nigeria. In an email on Feb 11th 2009, Mikhail Dombrovski hopes that “minor health issues” for George Ngoki from Nigeria, does not prevent him from attending their meeting. This quite likely refers to the outbreak of this new virus in western Africa. We claim that Dr. George brought the disease, originating from Nigeria as we said, with him and then infected Mikhail Dombrovski who he was buying weapons from, and possibly other people that he had contacts with in Dubai. All intercepted communication indicates that every other meeting planned for Dubai occurred in the week following this meeting between Mikhail and Dr. George. We thus claim that Mikhail then infected the other arms dealers and they infected all the customers throughout the week. This then led to the infection of all their home countries.

The order of spread of the disease is almost certainly clear from an analysis of the hospital admittance and death records during April, May and June 2009. The hospital admittance dates for patients in the outbreak were between mid April and the end of June with a peak around May 13th-21st. Dates of deaths across countries ranged from April 24th-May 4th to June 24th-30th with a global peak of May 22nd-28th. Although we had identified a set of 92 syndromes related to the disease in the dataset, further investigation narrowed this group into 39 distinct types of symptoms, with the most common ones being vomiting, diarrhea, abdominal pain, back pain, fever and nose bleeding. Having these 92 syndromes on average caused death 8 days after hospitalization. These syndromes clearly mark the general order in which countries experienced an outbreak of the disease by a few factors, including: the dates for first hospitalizations for the disease, first deaths caused by the disease, the jump in the number of deaths (onset of the disease), and the peak time of the disease. Ordering the countries by first recorded deaths caused by this disease, we have the following list: Kenya, Syria, Lebanon, Yemen, Pakistan, Saudi Arabia, Venezuela, Columbia and then Iran. Since all these nine countries were similar according to different factors in the outbreaks, namely high mortality rates (Mean = .097),  92 syndromes and the timing of outbreaks, we believe that they were all affected by the pandemic. Thailand and Turkey did not follow these same patterns and this is why we believe they were unaffected by the disease.

The main reason for our hypothesis that the arms dealers spread the disease is that there is also a correlation between this order of countries in the spread of the disease and the order in which people from these countries met in Dubai. After the initial meeting between the core group of arms dealers and the Nigerian, there were meetings with people from countries in the following order: Turkey/Syria, Kenya, Gaza/Lebanon, Pakistan, Yemen/Saudi Arabia, Iran, Venezuela and Columbia. For some of them there is evidence that they had flights scheduled to leave a couple days after and for the others we can assume they also would not linger in Dubai, but would instead return to their home countries quickly with the weapons they purchased. So, it is reasonable to think that the order, in which most of these people left Dubai, is the same as the order of meetings. Thus the fact that this order closely matches the order of the spread of the outbreak in the other countries strongly supports our claims that the arms dealer’s customers spread the disease.

In some cases it is a fact that those who planned to meet in Dubai died half a month afterwards, coinciding with the onset of the disease in their countries. We know from a newspaper report that Thabiti Otieno and Nahid Owiti from Kenya, who were supposed to be in Dubai on April 17th 2009, died in a hospital back in Kenya on May 3rd. Also, Saleh Ahmed from Yemen, who had planned to be in Dubai on April 19th, was close to death in a hospital back in Yemen, again back in his own country. In addition, Nicolai Kuryakin died in a hospital in Paris with the disease. We make the assumption this happened after everyone met in Dubai since we have no evidence he would be coming into contact with anyone else from western Africa.

We will now briefly outline the connections between arms dealer’s customers from each country and why they did or did not spread the disease.

A group of customers from Turkey and Syria referred to their group as a “school”. It was located in Syria, had many new “students”, and was in need of “textbooks” which they planned to buy from a professor from Moscow. At the same time, they ordered “farming equipment” over the phone and for this arranged for a large sum of money in Dubai during the week in April. While Syria was one of the first countries to experience an outbreak of the disease, Turkey was not affected at all. Since we have no intelligence on this area following April we can only make the hypothesis that this group only travelled to Syria in order to deliver their new purchase to their “school”. At this point it is possible that those from Turkey had died from the disease before they had a chance to return there.

From phone conversations and their likely history of arms dealing over the past year it is fairly certain that Thabiti Otieno and Nahid Owiti from Kenya attended the meetings in Dubai. As mentioned above they died at the start of May. From a variety of sources in Kenya, we are fairly certain these two are deeply involved with shipping weapons through Kenya to Sudan that originate from the core group. Thus they would have had much contact with them including Nicolai who clearly contracted the disease and died in Paris.

There are known terrorist groups in Karachi, Pakistan (the Lashkar-e-Jhangvi), and Gaza, near Lebanon (the Martyrs Front of Judea) that almost certainly had representatives purchasing weapons in Dubai during the middle of this arms week. They would have then returned to their home countries which were then affected by the disease.

Saleh Ahmed is a known arms dealer in Yemen and Saudi Arabia who was arranging to attend a meeting in Dubai in the middle of the week most likely with Mikhail Dombrovski and possibly Nicolai. Thus he would likely have contracted the disease and spread it to Yemen and Saudi Arabia. Also, based on the geography of the region he may have even travelled through Saudi Arabia to Yemen after the meeting.

While it is uncertain who exactly in Iran is connected to the arms dealing, someone at the residence of Sattari Khurshid was directed by someone in Ukraine to meet with Nicolai near the end of the week. Also, a group of people from Venezuela and Colombia planned to go to Dubai right at the end of this time period (April 22nd, 2009) to purchase weapons from Mikhail. They had arranged this through online message board posts and emails. After so many earlier meetings it is likely that both the Iranian people and those from South America came into contact with other people who already had the disease. This would then explain how the disease spread to these countries and why it peaked in them after all the other countries previously described.

As mentioned, Thailand did not receive the outbreak and we are aware of one key player, Boonmee, who is from there. We have no evidence on which to rely for a definite reason why Boonmee did not spread the disease to his home country however we can speculate based on his roles as an arms dealer and close associate of Nicolai that he did not return home after the Dubai meetings. First of all, we have no evidence that Boonmee was purchasing weapons in Dubai so he may not have the motivation many others had to deliver a product back to his home. Secondly, Boonmee at least twice attended conferences with Nicolai in the past year and so he may have travelled with him again after the Dubai meetings. We know that Nicolai died in Paris so perhaps Boonmee did as well.

Well many of these details in the above hypothesis fit together very well; there is a lack of intelligence on the whereabouts and condition of many of the key players in this scenario. To verify this is truly what happened, we recommend locating Boonmee Khemkhaeng, Mikhail Dombrovski and any of the others that surveillance was provided for before, in order to see that they are alive. It is likely that we will not find many of these people alive and that they actually were among the first to die from the disease within their country. Judging by how Nicolai was removed from a flight to Moscow, it is possible that Moscow and the Ukraine avoided a large outbreak of the disease, but we have no data on this.