Thursday, July 6, 2017

Cuts to Science Funding- An Impassioned Essay Against Them!

There have been multiple threats to science budgets in multiple countries in the recent past. The biggest threat looming is, of course, Trump. Trump's budget sought to cut funding to the NIH by billions of dollars. While this seems overall like a bad idea to a lot of people, it is hard for a layperson to understand how exactly this would affect the average Joe on a day to day basis. When I found myself wondering the same, a recent experience with an HIV database brought the situation to clarity.

Here is the Stanford HIV database. It is free- anyone in the world with an internet connection can access it and use it. It is the world's largest curated database on HIV sequences- all sequences ever published anywhere in the world make their way into the Stanford database, fitted neatly into the grand scheme, accompanied by neat little texts on how each sequence advances our knowledge of the virus.

So after you amplify your HIV genes (the genes that we like the most are Reverse Transcriptase or RT and Protease or PR. These are the genes against whose products many HIV drugs are targeted. So a drug that targets RT is called RTI, short for Reverse Transcriptase Inhibitor; one that targets PR is called PI, or Protease inhibitor). The virus, being the ultra mutable thing that it is, will try to mutate against either or both of these medications in a patient taking these medicines to escape their effects and proliferate. If it successfully mutates itself so that the drugs no longer act against it, it is said to have acquired drug resistance mutations.

The Stanford database is what helps a researcher or diagnostician figure out if the most prevalent type of virus in an infected patient has drug resistance mutations or not.

You can input your sequence (as I've shown below) and select whatever HIV drugs the patient is on.


And then the software gives you back something like this:


It tells you if there are any specific mutations that lead to drug resistance and how it might affect any of the medications that the patient is on. But this is the most basic utility. 

The database is something out of a fantasy novel- just like Hogwarts is so massive that nobody is ever supposed to know all its nooks and crannies, every turn brings up something new and marvelous, every day reveals new aspects.

For instance, you can also look at how prevalent certain mutations are according to the particular type of HIV that is found in a region, and what kind of medications the person is on.


This table above shows you the various amino acids that are possible at a particular site and how prevalent they are in different subtypes of HIV(A, B, C and so on. What does a subtype mean? Just like we all belong to the species Homo sapiens, but obviously look different based on whether we have Mongoloid, Caucusoid, Negroid etc features, subtypes A, B, C etc are the different faces of HIV in different parts of the world. There are clear genetic differences, and each subtype has multiple millions of members belonging to it. Subtype B is found in North America and Western Europe; Subtype C in most of Africa and Asia; Subtype A in Eastern Europe etc). 

It also tells me what is the proportion of finding this mutation in people who are RTI Naive and RTI-Treated (You know what RTI means. Naive means those who are not taking this medicine yet and have never been exposed to it, therefore providing no selection pressure for the virus to try to mutate against it). 

How is this information useful? 
For instance, look at picture 2 above. You see how it says 'Other Mutations' and lists out a bunch of things like V35T, T39E etc? 

How do we know if this V35T is important or not in our patient? We scroll down the page that is shown on picture 3 and come to this:


What this picture tells me is that in Subtype C, the proportion of people who are RTI-Naive and having V35T mutation is 92% (the little superscripted number above the T) and the proportion of people who are RTI-treated and having V35T is 91%. So V35T appears to be really common among those with Subtype C virus and therefore very unlikely to be something that actually causes drug resistance because it is found in high proportions among people who have never taken an RTI medicine. So I don't have to worry about my patient having V35T. Similarly, I look at the other mutations in my list and see which ones are unlikely to be appearing and therefore merit a second look.

For every single site on the RT protein the table lists out what amino acids are likely to be seen, and how prevalent they are, for every single subtype of HIV. You can also click on each underlined amino acid and get the primary sources of literature upon which this table is based. 

The amount of work that has gone into making this table is mind boggling. 

And we haven't even scratched the surface yet. The Stanford Database also has tools that tell you the fold decrease in the effectiveness of a drug in the presence of each drug resistance mutation, how the fold decrease changes if there are multiple resistance mutations in combination, how treatment should be changed based on what mutations there are and what you can expect to see in the patient over time.
The database also tracks resistance mutations across the world. One of the most important and challenging issues facing us right now is the problem of transmitted drug resistance- imagine there's a guy who is HIV positive and taking his medicines. Unknown to him, he has developed resistance against some of his medicines. He has unsafe sex and transmits HIV to his partner. Now the transmitted HIV will be the drug resistant version. When the partner starts the default first line medications, they will not work at all and the partner will start failing very soon. The Stanford Database tracks the levels of these transmitted resistant viruses across the world by linking up to the WHO surveillance system.  

Who made all this? How was this even done? Answer: much of it is through NIH funding. Someone in Stanford (Dr. Robert Shafer and his team) had this brilliant vision and the funding through NIH is what made it reality. 

This database is one of many that the NIH has developed and hosts free to the public. There are multiple such mammoth works that link clinical, systemic, organ-level, and molecular data. There are databases for various types of cancer, infectious diseases other than HIV, auto-immune diseases, databases linking environmental hazards to various health problems. These are more than just collections of raw data. They are magnificent works of art.  

Without funding, how will these function? A funding cut to NIH in the US affects the whole world. 

Sunday, July 2, 2017

Crashing down to earth

There's always a new way to be dumb. You think, man, I'm in my mid-30s, I've seen a bit of life, I think I can handle most things that life throws at me now and sure enough, you find a new way to prove yourself so utterly stupid or naive that you wonder, what the heck have I learned in all these years?


Moronicus indicus

This is the name that I give every dumb ass person driving on the wrong side of the road. And Indians are probably the most populous in this class of humans.

It could be a super-fast highway with cars speeding at >100km/hr. But no matter. To the intrepid M.indicus, this is all the more reason why he has to drive in the exact opposite direction, preferably in the middle of the highway.

BBMP recently constituted a ban on all liquor sales within 500m of a highway supposedly to prevent accidents. Drunken driving is one thing, BBMP. But if you were truly serious about lowering the accident and death rates on highways, please exterminate that pestilent subclass, so uniquely found in India, the Moronicus indicus.