|
|
Bioinformatics.org
|
![[?]](https://www.bioinformatics.org/images/icons/info.png)
|
|
Research
|
Online databases
Online analysis tools
Online education tools
|
|
Development
|
|
|
Forums
|
News & Commentary
Jobs Forum
(Career Center)
|
|
|
Java BioDataPlatform - Message forums
|
         
|
 |
|
Expanded view | Monitor forum | Save place
|
Message
|
|
Pals,
This project is initially refreshed in my mind while I was thinking of how to combine computers with biomedical research. At that moment, I was working purely as a Java software engineer, although I was from medical background.
I am working in a basic research lab in Philly now. Painful things I had never encountered began to bother me since I started to do research in "modern research lab". One thing, there are just too much redundance and too many little things to drag you out of your schedule. Is this part of reason that researchers are just working too hard to work things out? There are just too many spaces/holes left to get improved in basic research area. From order drugs to tailor them to fit your experiments, from management to do regular report, from data collection to analysis. Data overflowed everywhere, no good way to make them handle and search and do statistics to find out relationship in between...
The problem is in this biomedical arena nearly no common sense has been widely reached/accepted that biomedical data ( not just those gene array data) have come out to be such an ado and urgently need to be managed well and retrieved in a manner that can really help basic researchers/Ph.D students have a lot fun by marching to their targets in reasonable time frame.
What I proposed is just a frame, (not even a framework), since when we think that what we defined is very generic most likely ends up to very tight/restrictive for biomedical information system. Here what we need is nothing but participation from researchers. I am not talking about making them all to a programmer, but we need to define a easy to understand language-like way and load their "mind/thoughts" as configuration and build up a working platform for them to start.
This is tough! I know, that is why not me, a single dummy can do it. I am asking for feedback and comments from all of you. Tell me what is your pain when you doing research if you are from biomedical area. Otherwise, you'd better be ready to be a good architect to engineer it.
JBDP admin
Scott Li
|
|
Post a followup to this message:
You have to be to post a reply.
Thread view
|
|
Pals,
This project is initially refreshed in my mind while I was thinking of how to combine computers with biomedical research. At that moment, I was working purely as a Java software engineer, although I was from medical background.
I am working in a basic research lab in Philly now. Painful things I had never encountered began to bother me since I started to do research in "modern research lab". One thing, there are just too much redundance and too many little things to drag you out of your schedule. Is this part of reason that researchers are just working too hard to work things out? There are just too many spaces/holes left to get improved in basic research area. From order drugs to tailor them to fit your experiments, from management to do regular report, from data collection to analysis. Data overflowed everywhere, no good way to make them handle and search and do statistics to find out relationship in between...
The problem is in this biomedical arena nearly no common sense has been widely reached/accepted that biomedical data ( not just those gene array data) have come out to be such an ado and urgently need to be managed well and retrieved in a manner that can really help basic researchers/Ph.D students have a lot fun by marching to their targets in reasonable time frame.
What I proposed is just a frame, (not even a framework), since when we think that what we defined is very generic most likely ends up to very tight/restrictive for biomedical information system. Here what we need is nothing but participation from researchers. I am not talking about making them all to a programmer, but we need to define a easy to understand language-like way and load their "mind/thoughts" as configuration and build up a working platform for them to start.
This is tough! I know, that is why not me, a single dummy can do it. I am asking for feedback and comments from all of you. Tell me what is your pain when you doing research if you are from biomedical area. Otherwise, you'd better be ready to be a good architect to engineer it.
JBDP admin
Scott Li
|
|
|
|
|
|
Pals,
This project is initially refreshed in my mind while I was thinking of how to combine computers with biomedical research. At that moment, I was working purely as a Java software engineer, although I was from medical background.
I am working in a basic research lab in Philly now. Painful things I had never encountered began to bother me since I started to do research in "modern research lab". One thing, there are just too much redundance and too many little things to drag you out of your schedule. Is this part of reason that researchers are just working too hard to work things out? There are just too many spaces/holes left to get improved in basic research area. From order drugs to tailor them to fit your experiments, from management to do regular report, from data collection to analysis. Data overflowed everywhere, no good way to make them handle and search and do statistics to find out relationship in between...
The problem is in this biomedical arena nearly no common sense has been widely reached/accepted that biomedical data ( not just those gene array data) have come out to be such an ado and urgently need to be managed well and retrieved in a manner that can really help basic researchers/Ph.D students have a lot fun by marching to their targets in reasonable time frame.
What I proposed is just a frame, (not even a framework), since when we think that what we defined is very generic most likely ends up to very tight/restrictive for biomedical information system. Here what we need is nothing but participation from researchers. I am not talking about making them all to a programmer, but we need to define a easy to understand language-like way and load their "mind/thoughts" as configuration and build up a working platform for them to start.
This is tough! I know, that is why not me, a single dummy can do it. I am asking for feedback and comments from all of you. Tell me what is your pain when you doing research if you are from biomedical area. Otherwise, you'd better be ready to be a good architect to engineer it.
JBDP admin
Scott Li
|
|
|
|
|
|
|
BioData are not only limited to genome sequence analysis project, macromelecular structures, and functional experiments...There are just too many: different labs may apply their own setup and design their own project based what they have or what they prefer. Sometimes, some lab just devote their work to some areas and hard to/unwilling to change a lot in short term. In clinic, including some clinical data critical departments, they have their troubles to find the right way to organize their data...
We questioned many lab and people working in their lab, they were in trouble sometimes but find it is hard to find a good way to solve their problems, either because of their lack of knowledge of how to do right data handling in short time or they don't have proper support from software tools to help them...
What is in our mind now to have this project ready to go includes clearly discussion of following 4 parts:
1. GUI: their role, user case, the way to be dynamic.
2. Template Server: a server used/shared by a set of labs or institute that know their domain request/requriment. This server should be well-designed along business logic line and suppot dynamic GUI.
3. Business logic: logic of whole software. This is the layer that knows how to handle bioData logics.
4. Data: database and mining tools used for data management. Well-designed database is needed. More details would be discussed later with more other people with strong domain knowledge in biodata modeling.
Suggestions are welcome!
|
|
|
|
|
|
|
Database Design and Data Flow Plan:
In this project, to be able to have a flexible system, a well-designed database and data flow plan is required.
Database requirment: Database design must be able to adapt to new requirement from different user, who may came from different lab or even different group of researchers. As we have expected, bio-data varied as different research enviorment changed. Database would support wide range of users by managing and separating them into different group(s) or differnt classification.
Aslo, the complication involved in biodata need to be handled in a way that can communicate between different user and aslo the data set they need could be customized. Those can be supported by have "Dictionary Support" which would make taxonomy and mapping of data more efficient.
Third, database need to have flexibility to adapt user's change, which meaning, the whole design must be able to adopt this change. "Generic Propertie Table" and "Specific Properties Table" are designed to handle this. At the same time, they are more adpative to apply to data mining.
Data Flow Plan need to be considered as User Case in our situation, we pointed to the possible data flow for basic and clinical research and affairs: Basic User case is they can mining their data, they can searching data, they can use data set to do analysis and/or other useful work.
XML would be the best data media since we need to integrate user data flow with database and frontend presenation. XMl and Java support for it make it possible that system deals with all sorts of data object and using xml schemaa or dtd file as contact and/or a way of communication between user and our project. Precompiled XML to java Class would be used...Data communication and data flow between front end and middle layer logic (logic layer); and data flow between internal database and logic layer should be put into two layer. As we disscussed, we need to put an overall framewrok design later in this site
|
|
|
|
|
![[University of Birmingham]](https://www.bioinformatics.org/images/ack_banners/Birmingham-728x90.gif)
![[Patsnap]](https://www.bioinformatics.org/images/ack_banners/Patsnap-728x90.png)