Hidden Markov Models

By

Hidden Markov Models
Hidden Markov Models is one of the popular machines learning approach that can be used to decipher highly complex biological data.
One example would be CG-Island.

CG is the least frequent nucleotide in many genomes


Reason:
C within CG is easily methylated
Resulting methylation has tendency to mutate into T
However methylation is suppressed around genes in areas called CG-Island. Here CG appears relatively frequently.

Fair Bet Casino:
Through it CG-Islands can be modeled.
Dealer flips coin and player bets on outcome: heads or tails
Dealer uses either a biased coin (head with probability ¾) or fair coin head or tail equally likely
Given a sequence of coin tosses, problem to find out when dealer used a biased coin or a fair coin
Obviously, if you observe long line of heads it’s likely that dealer used biased coin and if you observe even distribution of heads and tails he likely used fair one.

Comments

Post a Comment

Popular posts from this blog

If all content (subject matter) are alike, how can we expect innovation?

By
Image
Problems with alike content: 1) Homogenization of skills When a finished product reaches the market, the product development cycle goes through the touch of many divergent skilled peoples. A product can have many components which require different skill sets of different people. For example, Website Development: When you begin to develop a website, you start with designing a prototype of how it will look, which requires the expertise of graphic designers, you need the skill of programmer to code for the website, you need to be a webmaster to host your website. If you look into the depth, there is much more complexity to it. Each website has its own requirements. A prototype can be designed by using different software, such as Inkscape, illustrator and many others. Each software requires a different kind of learning and experience. Similarly, from programmer side, which language you will be using, which framework, based upon your website requirements such as speed of development, scalab
Read more

No blackboard teaching please during Covid-19 shut down

By
Image
Coronavirus has created a major challenge in our education system. Schools are closed for many months, and teachers are finding ways to teach students remotely. But many are hopelessly failing at it, some think that it doesn’t work as there is limited evidence in support of remote learning in case of kids while others are trying to make it work.  Remote learning requires a different set of tools and different ways of handling students, but the first principles of pedagogy remains the same.  Old habits die hard. Many teachers started blackboard teaching through lecturing remotely through the zoom app or other apps like skype.  What is blackboard teaching? Blackboard teaching is when communication is one-sided with little student engagement. The teacher lectures the content and students listen. But there are serious problems with it.  Here are some of the compilation about patterns of bad teaching: Patterns for bad practices of teaching Remember, a well-designed video is many times super
Read more

The Competitive Collaboration Algorithm

By
Image
Competitive Collaboration Algorithm: It's a decentralized network where you compete to collaborate. Edge weight represents difficulty to get connected. Difficulty depends on factors such as distance, location, quality, price. We need to choose subgraph containing all the required nodes (e.g. nodes containing a,b,c,d ) that has a minimum weight or optimal weight. a,b,c,d can represent different services, whereas a1, a2 represents the same services from a different service provider. For example, 'a' can represent a student, 'b' can represent a teacher, 'c' can represent content provider such as a biology textbook and 'd' can represent a classroom or building. Similarly  'b1' represents teacher1, 'b2' represents teacher2, etc. So, For student1 (a1), the optimum subgraph or subgraph with minimum weight containing all the required services are a1,b1,c2, d2 with weight 0.5+2.3+4.2 . Usefulness of the model: 1) Prediction: One can make a pr
Read more