http://www.genomeweb.com/informatics/qa-stanfords-butte-says-omics-data-wealth-opportunity-bioinformaticians-medical-?page=show

have been looking about what you are doing… always wanna ask… but didn’t

Prof Daniel Chua first declared a ‘no fear zone’ for everyone becoz we are all coming from different background and to encourage discussion, no questions should be regarded as stupid. He also urged the participants to get involved in the discussion while the speaker is just there for providing background information. Instead of acquiring knowledge, everyone should be donors.

First speaker was Dr Robert Peckham (from the Dept of History, HKU) with a topic, Genomic Identities. He mentioned the research in genomics would change a biology lab to computer lab. I didn’t get too much from it his talk.

It was then followed by Dr Ron Zimmern from Hughes Hal, Cambridge, and PHG Foundation and gave a talk on topic, ‘Populations, Genetics, Risk’. He mentioned Nuffield Councils Intervention Ladder.

do nothing

provide information

enable choice

guide choice by changing the default policy

guide choice through incentives

guide choice through disincentives

restrict choice

eliminate choice

Population risk vs individual risk
He asked a few questions, but I can just copy three down:

extent to which it is considered legitmate for a state to attempt to change the behavior of its citizens to improve their health

whether any moral or social distribution exists between the four categories of public health activity

relevance and implication of the distinctions between a nudging & shoring and preferred preference and preference simp-liter

Dr Max Dentsch, Dept of Philosophy, HKU shared a talk on ‘Gene’ and Experimental Philosophy of Science. He split his talk into two main parts: what’s experimental philosophy (xphi)? and what’s applied to philosophy of genetics.
Traditionally philosophy requires one to think for most of the time and question what is intuitively true; whereas xphi must go out and ask people what they think about a case. it’s done systematically. for example, they found different biologists had no significantly different in defining what a gene is. However, when asked with ‘how a gene works?’ the answer varies with different training a biologist receives. This observation matters to interdisciplinary communication and how hard science is packaged for mass consumption. Moreover, xphi may be useful for understanding how different people understand certain key philosophy concept. This is different from psychology where xphi focus on how people think and psychology focus on what’s true about people’s thoughts.

Prof Pak Sham, Dept of Pycharity, HKU, lead everyone to think about “From genomics to humanity”. For example, can having understood how genes work in everyone lead to universal improvement, and can having understood how genes cause difference between people enable us to develop personal genomics. While we haven’t seen much benefits at the moment, we do recognize some of the potential misuse of the genomics information from our history. For example, Eugenics. Plato suggests “marriage number” where “the quality of the individual would be quantitatively analyzed, and persons of high numbers would be allowed to procreate with other persons of high numbers. In theory, this would lead to predictable results and the improvement of the human race.” Galton also observed low fertility among the intelligent. Pak suggested some possible safeguards, that include increasing public understanding, open & balanced debate, legislation and pointed out there is indeed limitation in the prediction of disease risk.

Personal Genomics is analysis of the DNA sequence of an individual to gain information that may be useful for improving the individual health or other benefits, where the levels of genetic analysis include heritability, gene finding and gene function, where the last one aims to find all specific sequence variants that contribute to heritability. Pak mentioned while we are having more understanding to our genome, we have only discovered a small percentage of genetics causes to our disease risk.

The discussion of genetics risk was brought further by next speakers, Prof Richard Fielding and Dr Dennis Ip, from School of Public Health, HKU, with a topic, “New Gene Blues: Fitting genetics to populations”. Fitting genetics for populations is to get genetic ideas from populations. But in any case, genes are not disease but different forms of a gene (e.g. different variants) would give risks or benefits to an individual given a particular environment. So, natural selection is on traits, not genes. A simple question like, ‘Is obesity a disease?’ gives a good illustration. Given high fat diet, a normal person with normally functional gene should get fat. It is the high fat diet (the environment) AND a normal body (genes) give rise to a ‘disease’, NOT a set of defective genes. In fact, population allele frequencies change very slowly. On the other hand, environments can and do change quite rapidly within a generation. The idea of personalized medicine is shifting the responsibility of risk to individual and ultimately bear the cost. More information provided implies no responsibility from the provider, even for drug. and because the responsibility is now shifted to the consumer, many different information are going to be distorted as long as people can make money on them. This may also lead to a shift of responsibility of the public health community from providing the public with epidemiology data, to accessing the validity of information in a market. The speaker ends the speech by asking the audience which way they prefer for their future, “Find your way” vs “Trust your guidance”. I think it’s very difficult to “Trust your guidance” when you know you can do better, but on the other hand, it’s also difficult to “Find your way” for all the ways in life. I think keeping things transparent is going to be more and more important when the world has more and more information, and then “Find your way” and “Trust your guidance” may both work fine :p.

Final talk was given by Prof Suet Yi Leung from Department of Pathology, HKU on “Genetic Testing for Hereditary Disease: Risk, Prevention, Self-Perception”. She shared her experience in using genetic information to help diagnose and then prevent patients from developing colorectal and endometrial cancers. As illustrated by many mendelian disease, genetic information does help in prevention and diagnosis, but how much it can be extended to complex disease is a question. She also shared her recent findings on ‘Heritable somatic methylation and inactivation of MSH2 in families with Lynch syndrome due to deletion of the 3′ exons of TACSTD1′ which has published in Nature Genetics 41, 112 – 117 (2008). I found this study particularly interesting, from which I may learn something to give future direction to my own PhD project.

by analyzing the citation pattern of a paper, we can estimate when a tool become popular, and which one is going to be out-of-date

—————
It is a note written on facebook on Friday, March 27, 2009 at 8:51am

I don’t think there is anything wrong the tab-delimited files. I think the usual genetic data files, .map and .dat are fine.

after I said, as a geneticist, one should not need to care about what data format a tool is supporting and goes straight to browse the pedigree and do analysis with the genotyped SNPs data.

Clear enough, one would not have any idea to improve on existing tool if one feels comfortable with the current situation. Maybe, at best, one would develop a tool which takes up data with one set of file format and ‘help’ to generate the other one thousands. but is that really a solution?

and happy enough, Google didn’t even feel comfortable with email, and then they invent Google wave:

I think all bioinformaticians should not feel comfortable with any of the current tools, we are too young to feel satisfied.

A: I am tenure professor and I want to apply a job at your institute.
I: I am afraid you’re over-qualified. You know, we have many high-school dropouts here working 48 hours a day and have published in Science or Nature, as first author, at their teenage. They dont need any degree to proof their ability or enhance their confident :p but we do count experience, what did you do?
A: well, I studied gene regulation of mainly gene1, gene2 and gene3. We used mouse as the model organism and created many different kinds of knockouts, knockins, etc. I know ChIP…
I: You did ChIP-SEQ?
A: What’s that?
I: ChIP coupled with next generation sequencing.
A: That’s too expensive to try.
I: really? but it’s too small in scale, dont you think? we do not even include it in our pipeline. We do collaborate with other researchers in this case, and help them with the sequencing part. Normally, we do it in tea time. 15 mins is about right.
A: er… I spent my whole life there…
I: really? how old are you?
A: I’ve got tenured two years earlier, so, (saying proudly) thirty-….
I: Oh! the average age of our group is 25, which is one year younger than that of NASA. You know, we are very proud of it… you are just … a bit over-qualified and …
A: over-aged…
I: well… our staff is just too brilliant
A: you can sequence them for the eliteome….
I: yes, we’re thinking of giving them their genome sequences as bonus… we will see…
I: in fact, we made all the data available with ftp.
A: Can I download 1000 Genomes from the ftp?
I: sure, if you dont mind… but we can ship the data to you in hard disk, I guess it’s faster; but we will charge you with the cost of shipping, labour, hard disk….
A: never mind, I will just ask my student to do the downloading…

*** The above conservation is not captured in a real life situation. It is just created by me after an inspiring talk given by sb from the institute. and I must say that this sb is really smart (say from heart). impressed by his presentation and his way of handling questions. ***

Let’s look into 1) first
Normally a PhD in States takes at least 5 years (mostly longer). So, we should instead ask if research in ‘gene network’ would be popular 5 years later. The short answer is yes, and the longer answer always has a ‘but…’ attached.

With a simple google on ‘gene network’, you can get 255,000 hints… surprisingly not a small number, when compared to 121,000 on GWAS ( genome-wide association studies); and by clicking into a few search result, for example:

http://genenetwork.sbl.bc.sinica.edu.tw/

you have an idea on what ‘gene network’ means.

Basically, with the advancement of biotechnology, we are expected to sequence all the genome in this world by 2030 (from recent talk by Prof. Peter Goodfellow). The next generation sequencing just opens up tones of opportunities to researcher to ask very specific research questions (e.g. which DNA sequence does my transcription factor bind to? rather than doing a motif search, you can design an experiment for it.) Moreover, enormous technologies are coming out with a very accurate measure of biological activities down to single molecules level. The promise of keeping track of a biological molecules across time allows researcher to really understand how the whole system works and give a more reasonable and testable model. Furthermore, the newly evolved linked structure of the web may finally provide a solution to handle the highly complex and connected biological data. All these made up the ‘yes’ component of the answer. I am reasonably optimistic that in 5 years time, we would study ‘gene’ as a network, no matter from a technical perspective or a market perspective.

As for the ‘but…’ part, there are mainly 3 concerns.
1. The development of the field is too rapid. Maybe what you used this year will be out-of-date next year. one needs to be very adaptive in learning new things and possibly suggests new methods given the existing ones. one also needs to be very quick in publishing your findings, or otherwise, the value will drop very fast.
2. There may not be many very experienced supervisors out there. As the previous generation scientists mainly educated with input->output unidirectional OR A+BAB bidirectional relationships of reactions. A network type of relations is often difficult to interpret and experimentally tested.
3. Data is not as clean as that in field business or physics. Biological data is often described with range or in relative to some controls. Assumptions in statistics (like independent data points) may not be valid. and the schema of describing a data often changed with the newly developed technology.

but I believe the above 3 are common to all fields under rapid development, and they are not lethal anyway .

and of course, even within ‘gene network’, you can choose from doing
1. developing physical or chemical methods to enable study of gene as network
2. using the existing technology to study gene as network
3. developing software to enable (2) to interpret results from (1)

different tasks would give one different challenges; but all of them should have a large amount of money running in; esp in the 21st century, we emphasize on quality of life, green energy and sustainability. So, if you are working either one of these, there should have jobs waiting for you. One’s capability is what only matters :p

and as for HK (I am trying to ask the 2nd part of the question), it is in fact a very simple city. If you are valuable, you can get a job. if you can get several ‘Nature’ and a few ‘Science’ in your PhD or post-doc, I dont care if you study shit or damn in the States, I am sure you can find a job here. So, don’t worry la…

and at the end of the day, I think one’s interest and passion should always be given higher rank; coz no matter which field one chooses, they are going to compete with sb who really love the job.

P.S. that’s the advice I gave to my friend. I intentionally post it on facebook to receive criticisms such that I can learn from others’ point of view and my friend also get comments from a more objective and diverse pool of people. You can find the discussion here.

I always think General Education is the best HKU offered to its students , and Prof Chan 陳龍生教授 is one of the best teachers in HKU. So, for every 《石頭記》 they co-organize, I will join . This time, we went to look for rocks in Zhaoqing, China. While most people shared their friendship-making experience, I want to share what I have learned in this trip.

The briefly session was held on 7 April 2009 and stated clearly the objectives of the trip:

• understand the particular uses of rocks in the past and at present
• understand how rocks have transformed communities
• appreciate that all items we use practically come from the Earth
• Gain Insights into problems and impacts caused by over-exploitation of rock resources

As for studying all subjects would finally arrive the study of philosophy, all researches or studies ultimately concern about human.

From understanding the particular uses of rocks in the past and at present, we know how people changed with their living style, what happened to a place and what was left after the resources were fully exploited.

From realizing how stone is polished, we learned how an industry was formed and how its development affected the local people, both the positive side and negative slide.

From experiencing the process of extracting rare materials, we learned that when something is valuable, people dont mind having a yield of 4g/ton.

All human activities appear to be a business decision. When there is market, people would try make it profitable and scarify those with less values. Every countries seem go through the same roadmap, but is there an alternative? mining is a business decision. can we increase the market value of ‘green product’? thereby changing the current business model? i dont think ‘greener’ process must cost more. and i dont think the current model is a sustainable one. and i dont think we can’t change.

The world has now changed from ’stone age’ to ‘digital age’. We can chase back the history (maybe 1,000 years ago) of one place or one group of people by studying the rocks there. However, can we also chase back what we have done before in the internet in our future? Internet has definitely changed our way of communication now. The journey to Zhaoqing has been extended for one week over the internet with many people doing the commenting and tagging. But, can it last? ICQ had been my first IM for keeping contacts with overseas friends, but all my friends migrated to MSN already and maybe soon to facebook. yet, it’s nothing but a 8-digit number now. it didn’t take even 10 years and it disappeared. will we be able to trace back our joy in the trip in our future?

A wise person knows:

1. When and how to make “the exception to every rule.”

2. When and how to improvise…. Read world problems are often irregular and ill-defined, and the context is always changing. A wise person is like a jazz musician, using the notes on the page but dancing around them.

3. How to use these moral skills in pursuit of the right aims. To serve other people, not to manuiplate other people.

4. ** most importantly, a wise person is made and not born. Wisdom depends on experience, and not just any experience. You need the time to get to know the people that you’re serving. You need permission to be allowed to improvise, to try new things, occasionally fail and to learn from your failures. and you need to be mentored by wise teachers…. that takes lots of experience… it takes lots of experience to learn how to care other people. 

I think the same goes for scientific research. When we gained the experience of having irreproducible experimental results, what we should do next is how we can ensure others could reproduce our result given the procedure we described and possibly reanalysis our data from a different perspectives. and not so far, we will end up having the idea of open science and linked data.

One
—

Scientist is more creative than artist
Artist draw things based on what they experienced, they just need to express their feeling
Scientist draw things that they never see nor experienced, but they need to do many experiments to prove their drawings

Two
—
Science is the pursuit of the truth, and there is one truth on the earth
So, without me, sooner or later, sb will find out the solution
That’s just a matter of time
Artist sells their feelings, and every feelings are unique
if you didn’t draw it out, no one can

Given one and two, it’s quite clear what the world is looking for