Wednesday, December 31, 2014

For A New Year

Here we are! The last day of 2014. Just another day for anyone who does not follow Gregorian calendar. Since my life is going to be fragmented into chunks of 365 days it seems to be an appropriate time for contemplation of the past 365 days. The reason why I am writing this up is to boast about the fulfillment of my last year’s resolution. I committed myself to write one blog per week. It was not perfect since I missed some weeks. But I ended up writing 50 blogs, thus averaging 1.04 blogs per week!

2014 was not a bad year. I got a feel of how serious science is getting done. I have identified some potential fields to focus on. Let me broadly say ‘Genetic and 3-D regulation of transcription’. Transcription is a fundamental cellular process. A cell is identified by the proteins and RNA it contains, which are nothing but products of transcription. But one can argue that it is a two-way relationship. Transcription is affected by many factors particularly the environment and also interaction with other part of genome, which a major difference in the gene regulation of many single celled primitive organisms and highly evolved metazoans. There is another component that contribute to identity both at cellular and organismal level. That is the primary DNA sequence of an individual, ‘the genetics’. How genetics modifies transcription and vice versa seems to be an interesting area of research. But I need significant improvement in certain skill set. First and foremost are computational tools, such as R or Python. I need extensive hands on experience in using computational tools to tackle actual biological problems. Second component is not skill set, but knowledge base. I need to know everything about these areas I mentioned above. A lot of reading and discussion is needed.

So here I am! Taking another set of resolution. To get to know my field inside out I have decided to go back to classics. The plan is to write/discuss about at least one classic paper per month. It will be mostly focused on genetics/molecular biology. I am taking this resolution with lots of trepidation since it is going to be very busy year for me, both at home and in lab. I do not know how much time will I have at my disposal.  But the resolution has to be executed! Thanks 2014 and welcome 2015!

Thursday, December 25, 2014

Who Will Solve Spooky Actions in Biology?

Biology is becoming more and more inter-disciplinary and quantitative in nature. Studies aimed to find out the role of protein X and Y in the biologic process Z was the kind of problems that dominated biomedical research in the 1980’s and 90’s.  But everyone was aware that such studies were far from being complete. It was analogous to blind-men describing elephant by touching its different body parts. Each molecule in a cell functions in collaboration with many other molecules. Many times the effect of a gene product will be on another tissues located at a different end of the body through endocrine system (For e.g. Prolactin produced in a part of brain called anterior pituitary exerts it main physiological function in breast of pregnant woman). In a way it is similar to the butterfly effect in Chaos theory, i.e. an apparently insignificant perturbation (such as flapping the wings of a butterfly) can result in much amplified final outcome (such as hurricane) in a different part of the world. Everyone appreciate the fact that to understand the biology of a molecule we should understand its impact at systemic level.

Studying the system is not a simple task. Generating, analyzing and interpreting the data become very complex. That obviously necessitates application of programming and computing. That is a different topic, but what made me think in this line now is a conversation I had with my PI recently. We were discussing the transcriptional regulation in response to specific signaling programs. How does different part of genome interact to elicit a precise transcriptional output? A change in one part of the genome result in a molecular event in a region situated at a far away part of genome. How does that happen? Are these interactions are predetermined or random?  At least we know that some of them are very specific. From where does this specificity come from? All the information is written in the primary sequence of DNA. How does DNA behave under certain physical constrain should be same as how a string with similar tensile strength and composition would behave in similar environment. If so, can we predict the bending and interaction of different part of genome using computer simulation? When you try to model the bending pattern of a string in response to different variables it would transform to a physics problem. Apparently, when neuroscientists tried to solve the neural circuitry they had to seek help from astronomers since that data structure could be deconvoluted using algorithm designed for studies in astrophysics. Which freaking biologist would know about it unless s/he kept an open eye for developments happening in other venues of science? In fact I think the nuclear space in a cell is a universe in itself. How do different molecular events are coordinated to bring out a precise cellular state at a given moment? How does a molecular event going on in one part of the genome causes a ‘spooky action' at a distant end? The answer, for sure, is not going to be solved by traditional molecular biologists.

Monday, December 22, 2014

Eat Junk Before Sunset; If You Have To

I briefly went through an article published on the link between obesity and metabolic syndrome with the time when we eat our food. The bottom line is that anything we eat after sun set is ‘fat’. Of course, that is an overstatement, but it roughly conveys the message. There is an important component in body metabolism called circadian rhythm (body clock). Based on the day-night pattern our body has developed rhythm for metabolic activities. It is expected that we are active during daytime and have most of the meals during that period. Our body prefers to give a rest to food metabolism during night. So if you start snacking around midnight, the body won’t be prepared for that. It may not wake of all the system to completely metabolize the food you eat, but it just store it as fat to convert it to usable ATPs at a later point of time. The study is done in mice. The study shows that even if the animals eat high fat/high-fructose (=junk food of humans) during the normal wake up time they can maintain a health body weight. There are different conditions they try in the study, but basic message is that if you can avoid late night snacking (or preferably, eat food a regular time and have an early dinner) you will be doing yourself a big favor by reducing the chance of metabolic syndrome, obesity and diabetes.

Sunday, December 14, 2014

Her(r) Einstein

Finished reading biography of Einstein by Walter Isaacson. What a beautiful life! He is an example of a pure genius. After getting to know his story many scientists I thought were geniuses are demoted to ‘scientists who are much smarter than me’. First and foremost, Einstein was a rebel. He found a peculiar joy in challenging he authority, whether it is the authority in science or politics. It is this rebelliousness that prompted him to challenge the century old dogma created by the founders of modern physics. He had a deep understanding of what was the existing knowledge gap. He was deeply philosophical, reading all major philosophical thoughts from Kant to Spinoza. It gave him clarity of thought. A clear idea about what is right and what is wrong (in his own mind). He was not afraid to make mistakes, and always willing to correct mistakes whenever it was evident to him, two important qualities needed for any scientist. He had a very romantic outlook. He loved nature and its deep-rooted mysteries hidden from ordinary minds.  A pure human being! After reading this book, I was a little disappointed because I realized that I could never be an Einstein (you might be thinking this:" We already knew it after reading all junk you have written here, you Einstein"). People like him are born, not made.

This week I also found time to watch ‘Her’ a very nice movie by Spike Jonze. It is a story of search for love and companionship in a near future world. The main character (Theodore) of movie is in the verge of a divorce and obviously is lonely. That is when a new operating system (OS1) was introduced. It has a very advanced algorithm and its response to the user evolve continuously from the feedback it gets from them. Theodore fall in love with the female voice of OS1 (Samantha) that he interacts it. Samantha also responds to cues from Theodore. In a way Samantha is nothing but the reflection of Theodore thoughts. The OS evolved to satisfy the expectation of Theodore and started responding accordingly. He therefore got a companion of his dreams. But OS1 started evolving beyond the world of Theodore. As Samantha says towards the end, that Theodore is a book that she loves very much. She does not want it to end, so she is reading it so slowly assuming that space between the words is infinity. Once you finish reading a book you do not want to keep on re-reading it. For various reasons Samantha leaves Theodore (and all others who was in love with OS1). Although that was a shock Theodore re-invent his concept of love. It is a painful learning process of Theodore, but something that would help him to face his fears, i.e. dealing with real emotions. When do you feel loved? I think it happens when someone get into our mind and feel the fears and pains we experience to lessen them and share our joys to amplify them.

Monday, December 8, 2014

Known Unknowns

I was reading an article about the experience of someone who sequenced their DNA to find out the mutations in genes that increase her chances of getting cancer.  Right now they report a panel of 21 genes that have shown a direct association with incidence of various cancer (Breast, Ovarian, Prostate). A known mutation is these genes can predict with high level of confidence about the chances of getting a cancer in ones lifetime.  If there is a mutation in 11 of them, you have high risk of having cancer. Mutations in 3 others are classified a moderate risk and rest are called ‘newer genes’. There are some obvious names in the panel. Tumor Protein 53 (TP53) for example, is a well-studied tumor suppressor gene. When a cell get rogue by starting to proliferate uncontrollably (i.e. turn cancerous) this protein get alerted and decides whether that cell has to commit suicide (apoptosis) or stop dividing by enter into a stage of arrested development. But if there is a mutation in this protein, it will not be able to do this job properly, resulting is propagation of rogue tumor cells and full-blown cancer. Therefore, as expected, mutation in this gene can cause almost 100 % chance of getting tumor in females and 73% chance of getting tumor in men.

Now, the next question is what can we do about it. If there is mutation in some specific protein, for example, BRCA1 in females, it significantly increases the chances of getting breast or ovarian cancer in them. So one preventive measure is to remove ovaries or breast in women with this mutation. Such ‘prophylactic’ removal of high-risk organs can reduce the chance of cancer occurrence by 95%.  This is a good thing. But what about mutation in proteins like TP53, which are indicated in cancer in many organs such as breast, brain, ovaries, soft tissues and many more. We cannot remove all these tissues fearing death by cancer. One possibility is to watch out for tumor. Death due to cancer can be significantly reduced if cancer can be detected very early on in its development. An alternated scenario is to correct this mutation in selected high-risk tissues. I.e. if there is a change in one letter (nucleotide) in DNA sequence, just fix it, similar to the spelling check application in MS Word work. There are multiple problems here, first and foremost, how to edit the wrong sequence? There are some ‘genome editing’ technologies (TALEN, CRISPR, etc.) that area coming up to address this issue. Second question is can we edit it in all the cells in all the tissues? Probably no. So where should we correct it. One possibility is to remove the whole tissue from your body (E.g. bone marrow, in case of leukemia) and replace them with corrected marrow that is synthesized from your own precursor cells for marrow. There are other alternatives such as corrected stem cell therapy or synthetic lethality approaches and few more, but may be a topic for another more technical blog. These may sound like science fiction. But it is becoming a reality with modern biomedical research.

Another point mentioned in the article I mentioned is about mutation of ‘Unknown Significance’. These are mutations (because it varies from vast majority of other individuals) but no studies have shown its association with any diseases yet. So we medical science cannot say whether this mutation put you in any danger yet. But future studies may show some correlation; some times there may not be any health impact for them. That is where the current sequencing result analysis stands. Well this is the risk assessment for cancer only. There are some other mutations associated with other health conditions such as Parkinson’s, Alzheimer’s disease, various heart and immunological disorders.  But again the catalogue is not very long.  What does it mean, there is still a lot of unknowns the genome. We still do not know what different mutations means. Our technological capability to sequence the genome has far out competed our existing knowledge base to properly interpret the data coming out of that technology. The sequencing machine to a geneticist is equivalent to complete works of Shakespeare to a four year old. But the medical field is catching up. How long will it take to completely assess the risk of individual mutations and other alterations in the genome? A good 50 years is my guess.

Inspired by the article ‘Unknown Significance’ by Couzin-Frankel, J (2014, Science).