Where to Start
Where is a good place to start looking at some of the tools that are posted? There are a lot of add-ins, scripts and files on the HTS tools and T-SQL pages. How do you get up and running as quickly as possible with everything that is presented?
The HTS.xla file is an easy add-in to get started with. It shows how to acquire some tricky data files, how to run some numbers through some functions and how to format the data back into a format that is similar to how the data originally looked. It is a great lead-in to understanding some useful concepts associated with relational database management systems (RDBMS) like Access or SQL Server.
If you are accustomed to using a RDBMS the Excel add-in is still a good starting point. You might be surprised at how easy it is to solve some problems with a "put anything anywhere" type of application. Of course you have to be careful to put things in the right place since proper organization is left up to the user in many cases.
After Excel, take a look at some of the Access add-ins. Start off with the SampleData.mdb file. This file has a module with a VBA procedure that will generate a table with some sample data. Once you have generated some sample data you can install the PercentControl3.mde add-in. This add-in calculates percent of control values generated off of the randomized data that was created from the SampleData.mdb procedure. Next try the ControlWellStats.mde to generate some statistical values. At this point the database will have several tables and you can start thinking about relationships and your logical model. The PatternMaker.mde helps with the process of normalizing data, which encompasses ideas about building a good logical model. To help make sense out of some of the statistical values that are generated take a look at the Predictor.mde. This is an assay calculator that pulls a lot of the statistical concepts together.
The RiseFallTrends.mde add-in breaks away a little from the other Access add-ins since it wasn't developed specifically for HTS. It helps to solve a problem that is fairly common, which is finding groups of rising and falling values in a column of numbers. It is a very worthwhile add-in to take a look at and fun to run some numbers through.
The scripts on the T-SQL page are very easy to run. The RandomDataScript.sql and SP_POC.sql scripts show how to solve the same sort of problems that are covered by some of the Access add-ins. This allows for an easy transition into using SQL Server if you are unfamiliar with this application. The UniquePairs scripts show two different coding and conceptual approaches to solving the same problem. Take a look and see if you can figure out how they work. They illustrate some important ideas about set based programming and T-SQL.
With T-SQL you are getting more into the logic and away from the Graphical User Interface (GUI) approach of hiding the underlying complexity. It takes a little practice to be able to understand how to manipulate data and databases using T-SQL scripts. T-SQL is a very powerful language though and very worthwhile to understand. Hopefully the scripts posted are interesting enough to get scientists started in using this powerful application.
I hope what is posted is useful for solving problems and also challenging enough to spark new ideas and interest in data analysis and HTS.
C. Eric Cashon
cecashon@aol.com.