Paul Boutin investigates. Color me sceptical. Yes, nowadays it is possible make synthetic genes. The usual length is about 1000 bp, about enough for one gene, or maybe half of one. There are so many problems to building a really deadly virus and distributing it. Let's think about it.

1) Making the virus, or obtaining it. You have to really know what you are doing, be a virology expert, or you will get seriously infected. Viruses like Ebola have a short incubation time (a week or so), and high mortality. Read all about it in Joseph McCormick (et al's) book: Level 4: Virus hunters of the CDC. This is a great book by renowed virus experts, they were at the forefront of many African disease outbreaks including Ebola and AIDS.

So the simple thing to do is to fly to Africa and obtain patient samples who are accutely affected, then issolate the virus and grow it up. Alas, this disease feeds on humans, and even the CDC couldn't grow it up in large titers for many years. So currently this is a huge stumbling block.

2) You could make the virus from the genome right? NO, not really, you still need a host and intact viruses. It takes life to create viruses, it's not something done in vitro currently. You need to actually have the enzymes and proteins that allow the virus the replicate, not the DNA or RNA.

3) Synthesis. Can you synthesize the genome of a huge virus like smallpox? Maybe in the future, currently this is a huge undertaking.

4) Release. This will always be a problem. The most lethal viruses have incubation times of decades, allowing them to spread silently. HIV comes to mind. If you release Ebola, you may kill 10, or a hundred people. But soon quaranteen procedures and the lethality of the disease will stop the spread. There has not recently been a huge outbreak of viral disease, such as smallpox, or Ebola. Even in primative Africa, thanks to organizations such as Doctors without Borders, or the CDC, primate quarantene procedures have been effective in greatly limiting outbreaks.

Bacteria like Anthrax seem like a good candidate, except that aerosol release is problematic due to the large amount of particles needed for infection (and that it's treatable with antibiotics). What about drug resistant pathogens? So far only evolution and natural selection has provided the means to confer drug resistance, a process which takes decades or years. In the lab, drug resistance markers have been created through decades of work, this is for penicillin derivatives (for organisms such as Anthrax). However, these bugs are still susceptible to many forms of penecillin, and new non-penicillin antibiotics. And bacteria are likely a poor candidate for biowarfare anyway. Viruses can survive much longer, and are more lethal.

Anyway, i'm sceptical, this is not a mechanism to kill thousands or millions of people. Evolution and current technology is against it. But as a weapon of terror, biowarfare can still be quite effective, witness the recent Anthrax releases.