It’s over! Finally.. I am a doctor! After months of hard work and days of thunder, preparing a presentation, after the final test, yesterday I was proclaimed doctor.
I have to thank my supervisors, my friends.. everyone!
Now only relax!!
At least until monday!
After 30 hours of computation the Big machine (30k € worth) spitted a .gz file of only ~440MegaBytes. I was suspicious about this size. It’s ok that .gz format compresses a lot, but this is too much. So I unzipped the file and after that I wrote a few lines of code to inspect this big file (1.5GigaBytes after decompression).
End… after 50 millions lines I noticed the file was incomplete!
Today I’ve runned another instance of the program, with a lot of improvements. Memory usage dropped! Not so drastically, but it is still a good news. Even time is employed much better. Only 30 seconds per compound instance! In this way my matrix will be ready by this night.
It’s been a long time since I posted here the last time. Not so much but I feel like I have to say something to keep my followers up to date with my work.
I’ve just finished a section of my thesis work. It is a part that can be exploded in several little parts (yeah, just like that principle.. divide et impera).
First of all I have to parse a tab-separated-values file that contains a certain amount of chemical compounds used on a particular cancerous cell line (MCF7). In this part the software goal is to search for duplicates in this list and to extract a subset of unique compounds. After that I query PubChem database to identify the little molecules and to retrieve additional information such as synonyms and SMILES formula. Then I operate a conversion on the SMILES formula to get a bitset format named fingerprint. Using Tanimoto/Jaccard similarity score I calculate the chemical structure similarity among the compounds. This last step generates a triangular matrix which I will use, along with other bio-genetic informations, to identify groups of pharmaceuticals that have the same effect and share some structural parts. But this is the next step.
During my work I had the opportunity to learn to use LaTeX, a great typesetting system. It is not so intuitive in the beginning, expecially if you’ve always been using word processors like MS Word or Keynote, but if you give it a try it will surprise you. It’s a very powerful way to produce scientific documentation. I’m using TeXShop and it’s pretty simple to learn to use.
Now I have to go to sleep because tomorrow starts the new section of the thesis and I want to be in a good shape! Goodnight!
I’d like to use an excerpt from a speech of Winston Churchill at the house of commons during the first year of world war II.
I started to think about it when I started to study Mathematic Analysis. I knew that it would not have been simple but, as usual, I wanted to face the problem head on. Hard work is always rewarded. I have argued today, the oral part of the exam was the hardest to study, but I made it.
Now a couple of days of freedom, and then I will start to work upon my thesis.
“I have nothing to offer but blood, toil, tears and sweat. We have before us an ordeal of the most grievous kind. We have before us many, many long months of struggle and of suffering.
You ask, what is our aim? I can answer in one word: Victory. Victory at all costs — Victory in spite of all terror — Victory, however long and hard the road may be, for without victory there is no survival.” W.C.
En Plein! 5 exams passed on 5 given!
3 times 30..
It’s been hard to stand, but now it’s a pleasure! Someone called me the “recordman”, and maybe I am, but the only thing I can think about now is a week of relax! Software Engineering is one of the hardest exams to prepare. Now I am watching a movie. Maybe later I will start planning my next european trip. I must also build this damn bridge connection at home. Then it will start again, 3-4 exams to give, lessons to be followed and a lot of study!
Now I know I can do it, as I used to say a long time ago… Impossible is nothing!