Tuesday, May 29, 2007

Resolving the panorama

This post is about image stitching methods used to make a panoramic image. Panoramic images have become important in the digital age. Initially, panoramic images were developed to increase the field of view on the photograph. In the digital age, because one cannot print out pictures with resolutions less than 200 dots per inch (explained here), the method to take print outs for posters is to take a number of photographs with at least 15% overlap and stitch them together later using some software. In order to take the individual pictures that make a panoramic picture, the best technique involves using a tripod so that the camera lens only moves on a sphere eliminating parallax error. In addition, the aperture and shutter speed should not vary between the various pictures. More tips on the techniques of panoramic pictures can be found all over Google or by sending an email to me. This post is more about the science behind stitching the images of a panoramic picture.

The idea of image stitching is to take multiple images and to make a single image from them with an invisible seam and such that the mosaic pictures remains true to the individual images (in other words, does not change the lighting effects too much). This is different from just placing the images side by side because there will be differences in the lighting between the 2 images and that would lead to a prominent seam in the mosaic picture.



This Figure shows 3 photos and the locations of the seams are shown in black boxes on each picture and the final mosaic formed from all three pictures.

The first step is to find points that are equivalent in 2 overlapping pictures [1]. This can be done by taking into consideration a certain amount of pixels in the neighborhood of a pixel from 2 pictures and finding the regions that overlap in colors between the 2 pictures. Then the images are placed or warped on a surface such as a cylinder (because the panoramic picture is a 2-dimensional representation of the overlapping pictures in a cylinder quite often). After this step the curve is found that gives the most amount of overlap between the equivalent pixels on both images. Then the images are stitched together with color correction. I will deal in this post with the various algorithms for color correction.



This figure is an example of the Feathering approach.


1. Feathering (Alpha Blending): In this method, at the seams (the regions of overlap), the pixels of the blended image are given colors that are effectively linear combinations of the pixel colors of the 1st image and the 2nd image. The effect is to blur the differences of both images at the edges. In this method, an optimal window size is found so that the blurring is least visible.




This figure shows the optimal blend between the 2 figures in the previous figure.


2. Pyramid Blending: In addition to the pixel representation of images, images can also be stored as pyramids. This is a data compression method in which a the image is stored as a hierarchy or pyramid of low-pass filtered versions of the original image so that successive levels correspond to lower frequencies (dividing the images into different layers that vary over a smaller or larger region of space so that the sum of it gives you the original image). During the blending method described above, the lower frequencies (which vary over a larger distance) are blended over spatially larger distance and the higher frequencies are blended over a spatially lower distance [1] causing a more realistic blended image to be formed. Here during the pyramid forming process, the 2nd derivatives of the images (Laplacian) are taken into consideration while forming the pyramid and the blended pyramid is formed and reintegrated to form the final blended image.



This figure shows the pyramid representation of the pixels in an image and the pyramid blending approach.

3. Gradient Domain Blending: Instead of making a low resolution mapping of the image as above, the gradient domain blending method requires the calculation of the 1st derivative of the images. Hence, the image resolution is not reduced before the blending process, but the idea is the same as above. This method is also developed to find the optimal window size for alpha blending and is adaptive to regions that vary fast or slower.


This figure shows the gradient blend approach.

Sources:
[1]: http://www.cs.huji.ac.il/course/2005/impr/lectures2005/Tirgul10_BW.pdf
[2]:
http://rfv.insa-lyon.fr/~jolion/IP2000/report/node78.html

Wikipedia article on feathering.

All figures taken from http://www.cs.huji.ac.il/course/2005/impr/lectures2005/Tirgul10_BW.pdf

Thursday, May 24, 2007

Biological Control - Doing it yourself.

It was not too long back that the whole of biology was very protein and DNA centric. The reasoning was that proteins were used to do all the work in the cell - be it chemical work (enzymes), or physical work (motors, and pumps). DNA was important because it provided all the information to make the proteins and contain the set of genetic instructions that are passed on from generation to generation. For long, there was a battle whether DNA was more important or proteins were more important neglecting DNA's chemical cousin RNA.

RNA was considered as a step required in modern organisms to convert DNA to proteins. RNA is made up of nearly the same chemical constituents as DNA but it is more flexible and can have wide ranging 3 dimensional structures unlike DNA's double helical structure. However, this increased flexibility comes at a price - RNA is more unstable and in modern cells, a single molecule of RNA does not remain functional for long periods of time (mean life time is approx 5 minutes in E.coli).

Of course, all this changed when it was found that RNA molecules could be used as catalysts and even in modern day cells, there are some RNA catalysts also called ribozymes (and the list of ribozymes discovered keeps increasing). RNA captivated the imagination of biologists as this was a molecule that could store genetic information as well as be used as catalysts - taking on the dual role of enzymes and information storage. All of a sudden, RNA was considered to be at the origin of life as we know it. However in the RNA world hypothesis, one should take into consideration that it is not that only RNA is present. It only postulates that RNA is present and is dominant but other biochemicals such as peptides (small proteins) and DNA oligomers (small DNA molecules) are also present and aiding life (idea originally proposed in [1]).

One of the biggest controversies against the RNA world hypothesis has been that it does not play that big a role in modern cells. However, it has been found more recently that there are many RNA control elements in the cell. One such control element is the riboswitch. For a gene to be made, the DNA gets converted into a message called the mRNA (messenger RNA) which later gets converted to the protein equivalent to that message. It has increasingly been found that mRNA do not contain only the message to be read but certain control elements could also be present in the mRNA. These control elements are called riboswitch.

Lets take an example. Supposing you want to make Vitamin B1. There is an intermediate in its biochemical pathway called thiamine pyrophosphate (TPP). TPP is also important for nucleotide (the chemical constituent of RNA and DNA) and amino acid (the chemical constituent of proteins) biosynthesis and is important for the cell to have the right amount of TPP channeled into the different biochemical pathways. When too much of TPP is present in the cell, TPP binds to a certain riboswitch in it's own biochemical pathway. This causes the riboswitch [2] to suddenly have a defined 3-dimensional structure (from an earlier random or semi structured RNA element). This defined 3-dimensional structure also blocks the production of the protein for making more TPP. The switch in the mRNA turns the production of the protein that makes TPP on or off depending on whether enough TPP is present in the cell or not - hence regulating the production of TPP itself. So far, riboswitches are found more in the microbial world and are only now being found in the eukaryotic world.

Now, in the latest issue of Nature, the first riboswitch that controls splicing in higher organisms such as fungus has been found [3]. Splicing is the mechanism by which parts of the mRNA are removed before the protein is made so that parts of the DNA never translated in the protein. Alternative splicing is the mechanism by which a single gene at the DNA level can be translated into multiple protein molecules. This is done by excising different parts of the mRNA (excising the DNA only in one situation and not another) before it gets converted to protein. Splicing and alternative splicing occurs only in eukaryotes and has also been discussed here.

Anyways, the first riboswitch in the mRNA have been found to function for alternative splicing purposes. The TPP biochemical pathway discussed above is the system that they found riboswitches in. In this case, when TPP was present, the riboswitch forms a three dimensional structure that avoids splicing and the protein that is formed can not make more TPP. So the objective was again control of TPP concentration in the cell but the means used was alternative splicing instead of just blocking formation of protein. The implications of these results will only come out with time, but there is speculation that this opens up a whole pandora's box on riboswitches that could be found in eukaryotes.

[1] The Genetic Code - Carl Woese, 1968.
[2] Thiamine derivatives bind messenger RNAs directly to regulate bacterial gene expression. Wade Winkler Ali Nahvi & Ronald R. Breaker. Nature 419, 952 - 956 (2002).
[3] Control of alternative RNA splicing and gene expression by eukaryotic riboswitches. Ming T. Cheah, Andreas Wachter, Narasimhan Sudarsan & Ronald R. Breaker. Nature 447:497 (2007) and its companion discussion article - Molecular biology: RNA in control. Benjamin J. Blencowe & May Khanna. 447:391 (2007)

pdf of all cited aritcles avaiable on request

Monday, May 21, 2007

Battle of sexes

Human Beings are diploid – that is each of us contains a copy of chromosome from our Mom and one from Dad. This gives us the advantage of having a spare copy of any given gene. However, there are certain genes that are "marked"in the embryo in such a way that either the Mom's or the Dad's copy is selectively silenced. The end result is that some genes in our body come with instructions attached, I am from Mom, Use only me! or vice- versa. The process that does this is called imprinting – either maternal ( for use only Mom’s gene) or paternal ( for use only Dad’s gene).

Why develop this curious phenomenon? On the surface it seems to be counter productive to us humans. If the marked/imprinted gene is defective then there is no working copy left since the silenced copy form the other parent can never be used. So why evolve such a complex yet dangerous mechanism? Since the process became known to scientists in the early 60s, several hypotheses have been put forth as to why this must occur. One of the most popular one shows the peculiar nature of inherent in a gene – its selfishness.

The Haig hypothesis is simple - it relates the development of a baby to the parent's inherent fidelity. The hypothesis, put forth by David Haig, predicts that Mom and Dad have different interests when in it comes to the development of their baby in any non-monogamous species, and hence imprint genes that are involved in growth of the embryo. Simply put both Mom and Dad fight a genetic war when it comes to the baby, more so if either one of them is prone to promiscuous!

Is there evidence for this prediction?
There is an excellent study done with the “deer mice” Peromyscus. This is a perfect species genus as we have both monogamous and polygamous species that can interbreed namely, P. maniculatus and P. polionotus. The females of the dark brown Peromyscus maniculatus species are promiscuous (babies within a single litter often have different fathers). Peromyscus polionotus, the sandy mouse, however pairs for life.

Check scenario one - Dad screws around but Mom is faithful.
In this case, the dad knows that the chances that all the offspring that she might carry are all his is slim. So he has to think of a way to make sure his baby grows faster at the cost off all other siblings and even mom.

This is exactly what you see when mate the faithful Peromyscus polionotus female with the
P. maniculatus male.The pups obtained are huge and mothers die giving birth.
Reason ? Well, the Peromyscus maniculatus dad has put a copy of a gene that ensures his baby grows faster since the female of his species is promiscuous. But the poor faithful Peromyscus polionotus mom is not used to playing this war and hence has no defense against the signals he is sending in. So the babies, prompted by Dad's genes grow unchecked, use up the mom’s resources and kill mom.


Check Scenario two - Mom is promiscuous but Dad is not.
The mom knows that since all the litter she carries has her genes, she can spread her genes in the population if she can restrict the growth of any one fetus, to conserve resources for her offspring with other males. So the genes she imprints will slow fetal growth.
That is what happens when you mate the promiscous P. maniculatus females with a steadfast Peromyscus polionotus male - you get tiny pups.
What happens? In this case, the mom is using her imprinted copy to slow down growth of the babies but the counterpart signal to grow is never received from the dad. The result is puny babies.

What if both parents are not promiscuous or vice versa?
The offsprings from a P. maniculatus cross or from a Peromyscus polionotus cross are healthy and similar in size. Reason? Each partner has co-evolved the defenses. In case of the promiscuous pair, the dad signals the babies to grow faster and mom to grow slower. In the other pair, each parent has the same vested interest in the offspring. End result is a normal sized litter.

What about humans?
So far about 80 of the 30,000 or so genes in the human genome are currently known to be imprinted. More importantly, most of these genes seem to play a role in directing fetal growth! And in the expected direction if humans were not considered to be monogamous-- genes expressed from the dad’s copy generally increase resource transfer to the child, whereas maternally expressed genes reduce it. So our genes behave much in the same way as the promiscuous mice! However there are imprinted loci are also implicated in behaviour/ neurological cases (Prader -Willi Syndrome) indicating that there is more to understand about this phenomenon.

More support for the theory comes from early indications that of very little imprinting in in fish, amphibians, reptiles and birds. Since the hypothesis states that imprinted genes are linked with acquiring resources from parents, this makes sense. But imprinting does also exist in seed plants where the endosperm tissue acts as the placenta to feed the embryo. Why this is the case is still unclear. There is lot of research that is ongoing and more that needs to be done. As molecular tools improve, we will be dissecting roles of imprinted genes much easily.

Ref:
1. Dawson, W.D. Fertility and size inheritance in a Peromyscus species cross. Evolution 19, 44−55.
2. Vrana Et. al., Genomic imprinting is disrupted in interspecific Peromyscus hybrids, Nature Genetics, 20, 362 - 365

Sunday, May 13, 2007

Global Warming Facts - Part 1

(I'm referring to news articles rather than scientific articles, and avoiding technical discussions in order to keep this article readable to everybody.)

If I told you that the Ganges and the Brahmaputra will both dry up by the year 2035, how hard would you laugh at me? Now, what if it was the world's leading scientific authority on climate change that told you?

I'm sure every one of us knows at least a little bit about global warming: that it is primarily caused by the greenhouse effect, and that greenhouse gas levels in the atmosphere have been rising because of industrialization and deforestation, that rising global temperatures will melt polar ice caps thus causing sea levels to rise, and so on. However, until recently, we've all been led to believe that we have a century or two to cut greenhouse emissions and quell the problem. The key phrase there is "until recently", because climate science has now progressed enough to tell us how bad the situation really is.

How bad will India be hit?
The first sentence of this article must have sent alarm bells ringing in your head. But a little thought will tell you why the Ganges will dry up, if not when: the Ganges, and indeed all perennial rivers in North India, are fed by glaciers in the Himalayas. As global temperatures rise, the glaciers receive snow later and start melting earlier, causing them to gradually fall back to the colder regions. This news article [1] in the Hindu has a detailed discussion about the effect of global warming on glaciers. The world's leading authority on climate change, the Intergovernmental Panel on Climate Change (IPCC), believes that all North Indian rivers will turn seasonal, and ultimately dry up by the year 2035 itself if global warming remains unchecked.

But there's more. Another news article [2] confirms our worst fears: inundation of low-lying areas along the coastline owing to rising sea levels; drastic increase in heat-related deaths; dropping water tables; decreased crop productivity are some of the horrors outlined for us. Falling crop productivity due to the change in the length of the seasons is of particular concern, because there is an acute shortage of arable land in our country. With the population still growing rapidly, and crop productivity dropping, combined with the fact that we are already facing a grain shortage this year and have been forced to procure from abroad, the situation appears dire.

Is it fair? The major contributors to the greenhouse effect thus far are the developed nations, and even on an absolute basis (let us not even go into a per-capita basis), India's contribution to global warming is very little. And yet, we will be among the first to suffer its effects, as the change in climate will decrease crop productivity near the equator but actually increase it in the temperate regions. Effectively, the third world has been offered a very raw deal: suffer for something you didn't do, and still bear the yoke of cutting emissions because, frankly, at this point our planet needs all the help it can get.

How high is safe?
Let us leave India's concerns aside for now, take a step back and look at the global picture. Global temperatures have risen about 0.6 C on an average in the past century. There is a worldwide consensus among scientific circles that the adverse effects of global warming will probably be manageable for a rise in temperature upto 2 C, but beyond that, melting ice caps, unbalanced ecosystems, drastically reduced crop yields, etc. will cause worldwide disaster of monstrous proportions. If I haven't painted the picture clearly enough for you, read this article [3] and this article [4] detailing exactly what countries like Canada and Australia can expect in terms of "disaster".

But, is this where you heave a sigh and think, if it takes a century for the temperature to rise 0.6 C, then we have plenty of time to remedy the situation before the rise reaches 2 C? Wrong. You see, there is a lag between the rise in greenhouse gases and the rise in global temperatures. Scientists give the analogy of heating a metal plate directly, and then indirectly, by placing a metal block between the plate and the heat source: when you place the block, it takes some time before an increase in temperature at the heat source affects the plate; at the same time, if the heat source stabilizes or drops in temperature, the plate will continue to increase in temperature for a while before stabilizing or dropping. Thus, the increase in temperature now is a direct effect of rising greenhouse gas levels sometime in the 20th century. We are yet to reap the effect of the carbon dioxide we are currently dumping into the atmosphere! And the fact is, the amount of greenhouse gases that have been going into the atmosphere has been steadily accelerating over the past century.

So, where should we hold greenhouse gas levels in order to hold the global temperature rise to 2 C? The answer cannot be explained in one sentence, because there is some statistics involved. We cannot accurately predict the temperature rise from carbon dioxide levels yet; we have to talk in terms of probabilities. A recent study by Meinshausen et al. [5] gives some startling numbers. This is actually explained in much simpler terms in this press article [6]. The gist of it is that, we are already past the safe limit! You see, the current level of greenhouse gases in the atmosphere stands at 459 ppm of carbon dioxide equivalent (the actual concentration of CO2, corrected to include the effect of other greenhouse gases). According to the Meinshausen study, if atmospheric greenhouse concentrations are maintained at 450 ppm, the probability of global temperature rise crossing 2 C reaches unacceptable levels (> 50%). The current EU target is 550 ppm - at that level, we will be looking at a rise of around 3 C! In other words, emissions across the world should already be decreasing, not increasing at an accelerating pace. Countries around the world should be spending a significant percentage of their GDPs to save the planet, but everyone seems reluctant to move.

Panels and Reports
I had mentioned the IPCC earlier. The IPCC was formed by the UN and has actually been around since 1988. Over the years, it has established itself as the world's leading authority on climate change. It publishes its findings periodically, the assessment reports published this year being the fourth set, and the most controversial one because it reads more like a disaster movie script than a scientific report. Actually, there had been protests over the previous report that the IPCC is being alarmist, and the UK government ordered an independent study be made (a committee was appointed, led by Nicholas Stern), and its findings were released at the end of October 2006. The Stern Review actually reported that the IPCC had understated the situation in the third assessment report. You see, climate science is far from exact, and the IPCC tends to err on the conservative side. There are already publications that say that the IPCC has been conservative even in the fourth report - read this news article [7].

Perhaps the most important thing that the fourth assessment report has accomplished is that it has finally laid to rest claims that global warming is a myth. Yes, until a few years ago, there wasn't even a global consensus on whether global warming is the fault of man, because the waters got muddied by studies that showed that greenhouse gases, while absorbing heat radiated by the earth, happened to reflect sunlight coming in, thus reducing temperatures. Further, it is believed that geologically, the world is headed towards an ice age. Increasing global temperatures were attributed to periodic properties of the Sun! Now, at last, all these speculations have been laid to rest, and IPCC has stated that there is a 90% probability that the phenomenon of increasing global temperatures is anthropogenic (caused by man), and primarily because of greenhouse gases - what we've suspected all along. India, too, has finally woken up to the threat, and has set up a panel [Citation needed] to investigate the specific effects of global warming on India over the next few decades, and what remedial measures are feasible. The panel is to be headed by Mr. Pachauri himself, the current head of the IPCC.

To be continued...
In the next part: The Kyoto Protocol, Emissions Trading, Extreme weather events, Bush-bashing, cows, bees and more!

References

[1] The Great Himalayan Meltdown
[2] Climate Change Will Devastate India
[3] Dire consequences if global warming exceeds 2 degrees says IUCN release
[4] Two degrees of separation from disaster
[5] M. Meinshausen "What Does a 2 C Target Mean for Greenhouse Gas Concentrations? A Brief Analysis Based on Multi-Gas Emission Pathways and Several Climate Sensitivity Uncertainty Estimates." in H. Schellnhuber, et al., eds. Avoiding Dangerous Climate Change (Cambridge University Press, New York, 2006)
[6] The rich world's policy on greenhouse gas now seems clear: millions will die
[7] Some scientists protest draft of warming report

Wednesday, May 09, 2007

Attention Concerned Scientists in IN, KY and OH

If you are a scientist in Kentucky, Indiana or Ohio and are concerned about the scientifically inaccurate materials at the Ken Ham's Creationist museum, please sign this.

Statement of Concern
We, the undersigned scientists at universities and colleges in Kentucky,
Ohio, and Indiana, are concerned about scientifically inaccurate materials at the Answers in Genesis museum. Students who accept this material as scientifically valid are unlikely to succeed in science courses at the college level. These students will need remedial instruction in the nature of science, as well as in the specific areas of science misrepresented by Answers in Genesis.

Via Pharyngula

Tuesday, May 08, 2007

Is there anybody out there?

This piece in November promised a lot more and I have failed to deliver on my promise so far. This is my first attempt to catch up with what I had promised. This post will deal with the chemicals that one finds in asteroids that land on Earth and with it questions the possibility that the raw materials for life on Earth could have started by the availability of these chemicals and also discuss the possibility of panspermia.

First, I will start with astronomical spectroscopy. This is the method by which chemists identify the compounds present in space. When light or any electromagnetic radiation is passed through a sample, the sample absorbs and emits certain wavelengths of light better than others and the wavelengths that are emitted and absorbed can be used as a fingerprint analysis of the chemical nature of the compound itself. Unfortunately, the science is not as simple* as I mention here but will suffice for the discussion that follows.

There have been many meteors that have hit the Earth's surface and some of these impacts have been seen as the reason for major climate change in the Earth. The interest in astronomical spectroscopy was purely to understand the physical and chemical nature of the universe around us. But as soon as astronomical spectroscopy developed into a reliable science, it stood to reason that it could lead us to understand how life on Earth originated and whether there are traces of life elsewhere in the universe. Afterall, if life evolved on Earth, the chemicals responsible for life should have been present on Earth before that (and possibly elsewhere) and hence the chemical nature of these meteors became important to biology as well, but all these studies have not been localized to the meteors alone.

The interstellar medium is divided into the dense and diffuse kind. The diffuse interstellar medium is cold and icy material that is not too dense and is made up of neutral and charged ions of compounds of C, H, and N, and also contain compounds such as naphthalene, which are aromatic compounds. In the dense interstellar compounds, the temperature is close to 10K to 200K (freezing point is 273K) important compounds such as hydrogen gas, carbon monooxide, water, carbon dioxide, methane, methanol, ammonia and hydrogen disulfide were found among others. That is, it has a source for H, C, N, O, and S. Later, in some clouds they have also found organic acids and higher alcohols such as ethanol (pure delight!).

The meteorites that have hit close to home were found to be quite rich in the lower and higher organic compounds of the classes mentioned above but were also found to have trace quantities of natural as well as unnatural amino acids (natural defined as biologically natural), purines, and pyrimidines (the base compound in DNA and RNA). In addition trace quantities of phosphonates and other P containing compounds were also found (also found in DNA and RNA). What was also interesting is that some of these amino acids was found to be chiral in nature (like in biological systems). In other words, there is a way in space to make optically active compounds and not synthsize all the isomers in equal quantities. It is actively debated whether these meteors were contaminated by biologically active components on their way to the ground even though there is evidence that says that it was not contaminated.

To summarize, the raw materials for life to start could be found in meteors and other components of space and indeed, these compounds under the right condition could lead to life anywhere. I will deal later with attempts by scientist these days to understand how life started from these raw materials.

I would like to end with panspermia and I think wikipedia has a good definition - Panspermia is the hypothesis that "seeds" of life exist already in the Universe, that life on Earth may have originated through these "seeds", and that they may deliver or have delivered life to other habitable bodies.

It is kind of a whacky theory and people either do not believe it or do not want to believe it because it is a theory like intelligent design - once you have said it, there is no way to prove it right or wrong. It is a theory which is unscientific in nature. But one of the leading scientists believing in the theory was none other than the Nobel Laurette - Francis Crick. Finding these organic chemicals in space has only led to more evidence for this hypothesis.

* Before one performs spectroscopy of a sample, one has to attempt to purify all the compounds present in the sample which is not an easy job because the chemical nature of the substance is an unknown at the beginning. A variety of chromatographic techniques are used for this. In addition, even after the spectroscopy of the individual samples are performed, it does take some time to realize the exact chemical nature of the substance being examined.

References:
Wikipedia as usual - on Origin of life and Astronomical Spectroscopy and Panspermia.

Extraterrestrial Organic Matter: A review - William M. Irvine - Origins of Life and Evolution of Biospheres - Volume 28, 1998, 365-383.**
** I can provide the pdf of this document on request.

Monday, May 07, 2007

Starting with the parts and ending with the whole

The focus of today's blog post is going to be statistical mechanics. In statistical mechanics, one starts with the properties of the atoms/molecules/ions in a system and try to understand how the whole system will behave. So, one starts with the microscopic properties of the components of a system and tries to understand the macroscopic properties of the whole system. The macroscopic properties here implies properties such as volume or temperature or pressure that one measures in an experiment physically. I will provide an example to make one understand how difficult this task is:

Now lets consider a box or cylinder filled with gas molecules. As each individual gas molecule is small in volume, to fill up the whole container, one would require a very large amount of gas molecules, lets just say, something in the order of a mole (A mole contains 6.023 * 10^23 molecules of the gas. This might sound enormous but is actually a small number in terms of molecules. To place things in perspective, 1 mole of water is contained in only 18 grams of water, and a liter of water typically contains 55.556 moles of water). Let us assume for simplicity that these molecules obey Newton's laws of motion and do not undergo any quantum effects.

Even under these conditions, the molecules are all moving and the total energy of the system would be the sum of each molecule's individual kinetic energy and potential energy. In addition, there will be forces acting on each molecule due to the neighboring molecules as well as the ends of the container. So, by Newton's law of motion, each particle will have a unique acceleration induced on it and the position and velocity of each particle continuously changes within the box. It becomes a hopeless situation to even try to follow an individual particle's position and velocity as the position of the other particles affect the potential energy and force of the particle we are interested in.

Hence, what one does is try to come up with a probabilistic approach as to how the system's macroscopic properties are affected by it's microscopic properties. Most of the theory that is dealt with in statistical mechanics are valid only when there is a sufficiently large number of particles (as the derivations that will come up in the coming weeks will show) and will not hold true under other conditions. Using statistical mechanics, one can go beyond the simple Newton's laws of motion and try to derive/explain the laws of thermodynamics that one can measure experimentally.

Books to understand Statistical Mechanics:
Chandler, David (1987). Introduction to Modern Statistical Mechanics.
McQuarrie, Donald (2000). Statistical Mechanics.
R.K.Pathria (1996). Statistical Mechanics.

Book to understand Thermodynamics:
Above books and
Callen, Herbert B (2001). Thermodynamics and an Introduction to Thermostatistics.

Some of the above discussion was also inspired from the Wikipedia article on statistical mechanics.