Opasnet base structure

From Opasnet
Jump to: navigation, search



This page is about the old structure of Opasnet Base used in ca. 2008-2011. For a description about the current database, see Opasnet base 2.

Question

Structure and connections (lines) of the tables (boxes) in the Opasnet Base. All table identifiers are called id (so they can be called by like obj.id). When obj.id is referred to in another table such as actobj, it is called actobj.obj_id. The latter end of a one-to-many relationship is marked with a ring. Important substantive fields are listed inside table boxes.

Opasnet Base is a storage and retrieval system for results of variable and data from studies. What is the structure of Opasnet Base such that it enables the following functionalities?

  1. Storage of results of variables with uncertainties when necessary, and as multidimensional arrays when necessary.R↻
  2. Automatic retrieval of results when called from Opasnet wiki or other platforms or modelling systems.
  3. Description and handling of the indicess that a variable may take.
  4. It is possible to protect some results and data from reading by unauthorised persons.
  5. If is possible to build user interfaces for easily entering observations into the Base.

Answer

Opasnet base is a MySQL database located at http://base.opasnet.org.

Data structure

Main article: Data structures in Opasnet

All data should be convertible into the following format:

Observation
Year Sex Age Height Weight Description
2009 Male 20 178 70 An optional column for descriptive text about each row.
2009 Male 30 174 79
2010 Male 25 183 84
2010 Female 22 168 65

where

Names of explanation columns, also known as indices.
Explanation data, also known as locations. You can use these columns as search criteria.
Observation index, typically called "Observation". Common name for all observation columns
Names of observation columns. These are the parameters of interest.
Observation data. These are the actual measurements.

Table structure in the database

All tables

act
Uploads, updates, and other actions
Field Type Null Extra Key
id int(10) unsigned NO auto_increment PRI
acttype_id tinyint(3) unsigned NO MUL
who varchar(50) NO
comments varchar(250) YES
time timestamp NO
temp_id int(10) unsigned NO MUL
actloc
Locations of an act
Field Type Null Extra Key
actobj_id int(10) unsigned NO PRI
loc_id int(10) unsigned NO PRI
actobj
Acts of an object
Field Type Null Extra Key
id int(10) unsigned NO auto_increment PRI
act_id int(10) unsigned NO MUL
obj_id int(10) unsigned NO MUL
series_id int(10) unsigned NO MUL
unit varchar(64) YES
acttype
List of action types
Field Type Null Extra Key
id int(10) unsigned NO auto_increment PRI
acttype varchar(250) NO UNI
cell
Cells of an object
Field Type Null Extra Key
id int(12) unsigned NO auto_increment PRI
actobj_id int(10) unsigned NO MUL
mean float YES
sd float NO
n int(10) NO
sip varchar(2000) YES
loc
Location information
Field Type Null Extra Key
id int(10) unsigned NO auto_increment PRI
std_id int(10) unsigned NO MUL
obj_id_i int(10) unsigned NO MUL
location varchar(100) NO
roww mediumint(8) unsigned NO
description varchar(150) NO
loccell
Locations of a cell
Field Type Null Extra Key
cell_id int(10) unsigned NO PRI
loc_id int(10) unsigned NO PRI
obj
Object information (all objects)
Field Type Null Extra Key
id int(10) unsigned NO auto_increment PRI
ident varchar(20) NO UNI
name varchar(200) NO
objtype_id tinyint(3) unsigned NO MUL
page int(10) unsigned NO
wiki_id tinyint(3) unsigned NO
objtype
Types of objects
Field Type Null Extra Key
id tinyint(3) NO PRI
objtype varchar(30) NO
res
Result distribution (actual values)
Field Type Null Extra Key
id bigint(20) unsigned NO auto_increment PRI
cell_id int(12) unsigned NO MUL
obs int(10) unsigned NO
result float NO
restext varchar(250) YES
implausible binary(1) YES
wiki
Wiki information
Field Type Null Extra Key
id tinyint(3) NO PRI
url varchar(255) NO
wname varchar(20) NO

Contents of selected tables

Table objtype
id objtype
1 Variable
2 Study
3 Method
4 Assessment
5 Class
6 Index
7 Nugget
8 Encyclopedia article
9 Run
Table acttype
id acttype
1 Start object
2 Finish assessment
3 Update formula
4 Upload data (replace)
5 Upload data (append)
6 Review scope
7 Review definition
8 Add object info

Rationale

Data

Software

Because Opasnet base will contain very large amounts of mostly numerical information, the state-of-the-art structure is a SQL database. Because of its flexibility, ease of use, and cost, MySQL is an optimal choice among SQL software. In addition to the database software, a variable transfer protocol is needed on top of that so that the results of variables can be retrieved and new results stored either automatically by a calculating software, or manually by the user. Fancy presenting software can be built on top of the database, but that is not the topic of this page.

Storage and retrieval of results of variables

The most important functionality is to store and retrieve the results of variables. Because variables may take very different forms (from a single value such as natural constant to an uncertain spatio-temporal concentration field over the whole Europe), the database must be very flexible. The basic solution is described in the variable page, and it is only briefly summarised here. The result is described as

  P(R|x1,x2,...) 

where P(R) is the probability distribution of the result and x1 and x2 are defining locations of an index where a particular P(R) applies. Typically locations are operationalised as discrete indices. A variable must have at least one index. Uncertainty about the true value of the variable is operationalised as a random sample from the probability distribution, in such a way that the samples are located along an index Sample, which is a list of integers 1,2,3...n, where n=number of samples.


Dependencies

Replacing some cells

It is possible that there is a large data, where there is a need to update only a few cells while all others remain the same. How should this be done? There are a few potential alternatives.

  1. Use the current replace functionality. Replace all cells but most of them with the original value.
  2. Use a new act_type that is similar to the current append functionality. This should be understood in a way that if there are two (or more) identical cells (based on cell indices and locations), then the newest result is used and all older ones are discarded. (If the old append is used, then new info is just seen as a new row in the data table, not a replacement of an existing row.
  3. Add a new field into the cell (?) table with an updated cell_id (in a similar way than act_id and series_id are used in the actobj table). This way, the new cell can automatically inherit all locations of the old cell.

Formula structure

Now it has become clear that it is not enough to have samples of the result distributions. It must be possible to completely recalculate the result based on the information in the Opasnet Base. There are different approaches:

  • Calculate the result based on a formula that may refer to other variables called parents. This is a deterministic approach.
  • Calculate the result based on the marginal distribution and (conditional) rank correlations with parent variables. This is a probabilistic approach.

This approach requires new tables, namely Formula and Language.


11: . Do we need tables DIF and DIP like Uninet? --Jouni 21:50, 30 December 2009 (UTC) (type: truth; paradigms: science: relevant comment)
  • DIP
    • DIP_node_id
    • DIP_parent_node_id
    • DIP_corr_coeff
    • DIP_parent_index
  • DIF
    • DIF_node_id
    • DIF_formula
    • DIF_varnames_in_formula

Universal Opasnet Base

The idea of universal Opasnet Base says that it should be possible to store results in such a way that the results themselves are public but their interpretation is limited. For example, patient symptoms and clinical test results should be openly available for research, but information about whose results they are should be private. This can be achieved with the following database structure.

Universal Opasnet Base has some parts that exist in different versions depending on the privacy level. The yellow areas are e.g. a public area and a private area. The parts that are white are public.

Let's say that it is enough to have two security levels, public and private. A person wants to record personal health information into the database. She logs in with her personal user name. The private profile gives the name (say, Liisa) and social security number of the person, while the public profile says only "30-40-year-old woman in Finland". Liisa writes down her symptoms or medical information and saves them. This is what is stored in the databases:

Information stored in the public and private databases. The private database can read tables from the public one but not vice versa.
Table, field Private database Public database
act.who Liisa, 010175-1024 Woman, 30-40 a
act.when 2011-03-09 22:09:10 2011-03
obj.name N/A. Data is taken from public side. Pregnancy test
loccell.loc_id (locations and indices explained) Person = 010175-1024
Time = 2011-03-09
Test = Clearblue digital test
Age = 30-40
Sex = Female
Country = Finland
Time = 2011-03
Test = Clearblue digital test
res.restext N/A. Data is taken from public side. Pregnant 1-2 weeks.

Based on the information, anyone can see that there is a woman in Finland who has used a Clearblue pregnancy test and the result was positive. But there is no way an outsider could connect this information to any particular person, because all information that could be used for linking is located in the private website. However, an authorised person from health case could see the data in the private database and connect Liisa and the test result.

See also

A basic query for retrieving the full result of a variable upload (an example)

{{#sql-query: SELECT obj.ident, obj.name, obj.unit, obj.page, obj.wiki_id, comments, mean, sd, n, location, ind.ident, obs, result, restext FROM obj LEFT JOIN actobj ON actobj.obj_id = obj.id LEFT JOIN act ON actobj.act_id = act.id LEFT JOIN cell ON cell.actobj_id = actobj.id LEFT JOIN loccell ON loccell.cell_id = cell.id LEFT JOIN loc on loccell.loc_id = loc.id LEFT JOIN obj AS ind ON loc.obj_id_i = ind.id LEFT JOIN res ON res.cell_id = cell.id WHERE obj.ident = "Op_en1912" AND actobj.series_id = 190 LIMIT 0,100 }}

Some useful syntax