Frequently
Asked Questions
This site will answer commonly
asked questions about POPGENE, its application, and use. It will
be updated as frequently as warranted. POPGENE
Users should print
this section because it contains many useful answers!!!
Table of
Contents
How do I ...
including downloading, registration, & running POPGENE?
Where can
I find ... ?
Why doesn't
... ?
Who is ...
?
What is/are
....including commands & error messages when running POPGENE?
When is/will
... ?
How
do I get a copy of POPGENE ?
You can Download
POPGENE from this website.
Remember to register because
it will allow us to contact you electronically concerning future updates
etc.
You should also Download
the "Quick User Guide".
However, if your have previously
registered, you do not have to register again each time you download an
upgrade version of POPGENE. To download without registering, do the following:
1. On POPGENE Home page, click
on "Download latest version"
2. You are in a new page where
you will see "Please Register First!". Don't type your name and
e-mail address. Just click on the "Submit" button.
3. You are now in a new page.
Click on the "No - return to >>> Download Area <<<"
button.
4. You are now in the "Download"
page. Follow the on-screen instructions.
How do I install POPGENE
?
Start Microsoft® Window
(Windows 3.11, Windows 95 and Windows NT).
Go to the directory where you
downloaded POPGENE.
Type or double click on the
file you saved.
Follow the instructions on
your computer screen.
Go to the directory where you
have installed POPGENE and double click the POPGENE icon
to start the program.
How do I uninstall POPGENE
?
32- bit version
Start Microsoft® Windows.
Select "Start" menu,
"Settings", then "Control Panel".
Select "Add/Remove Programs"
and double click on "Population Genetics Analysis (32-bit Version)".
Follow the instructions on
your computer screen.
16- bit version
Start Microsoft® Windows.
Select PopGen16 folder.
Double click on "Uninstall
PopGen16".
Follow the instructions on
your computer screen.
How do I analyze RAPD data
using POPGENE ?
RAPDs are predominantly dominant
markers so that we cannot distinguish between the 1/1 and the 1/0 RAPD
phenotypes in diploid tissues. Procedure to estimate RAPD frequencies is
straightforward if the study populations are in Hardy-Weinberg Equilibrium
(see Clark and Lanigan in Mol Biol Evol 10:1096-1111, 1993). However, most
populations under investigation are not in HWE.
POPGENE enables you
to estimate population genetic parameters under Hardy-Weinberg Equilibrium
(HWE) and Hardy-Weinberg Disequilibrium (HWD). When you have estimates
of departure from HWE (i.e., Fis values)
for your populations, for example from previous study of co-dominant markers
such as isozymes, you can input the Fis values
into your data input file. Then, specify "Hardy-Weinberg Disequilibrium"
on the POPGENE menu when you analyze the data. POPGENE will
now estimate RAPD frequencies using the algorithm given by Chong, Yang
and Yeh (see Current Genetics 26:374-376, 1994). If you only have Fis
values for some populations, those populations without Fis
values are automatically assigned a Fis value
of zero.
Notice that if you have Fis
values in your data input file but you specify "Hardy-Weinberg Equilibrium"
on the POPGENE menu when you analyze the data, POPGENE will
disregard your Fis values and all populations
are automatically assigned a Fis value of zero.
Notice that you may examine
the effect of departure from HWE on population genetic estimates in your
populations by comparing the results based on HWE with the results based
on HWD. Even if you do not have Fis values for
your populations, you probably can use information pertaining to their
population size, age, degree of isolation etc.
How do I interpret the POPGENE
output for "Neutrality" test?
You compare the "Obs.
F" to "L*95" and "U*95", which are respectively,
the lower and upper 95% confidence interval. If "Obs. F" is within
this confidence interval, the locus is neutral; otherwise, it is not.
How do I cite POPGENE
?
We will submit POPGENE
to the Journal of Heredity in 1998. At this time, you can cite POPGENE
as:
YEH, FRANCIS C., YANG, R-C.,
BOYLE, TIMOTHY, B.J., YE, Z-H., and MAO, JUDY X. 1997. POPGENE, the user-friendly
shareware for population genetic analysis. Molecular Biology and Biotechnology
Centre, University of Alberta, Canada.
Some users prefer citation
out of a journal, and for that, you can cite POPGENE as:
YEH, F.C. and BOYLE, T.J.B.
1997. Population genetic analysis of co-dominant and dominant markers and
quantitative traits. Belgian Journal of Botany 129: 157.
Where
can I find more about POPGENE if I have questions?
You have at least four options:
Read the POPGENE "Quick
User Guide".
Contact the POPGENE
staff (People).
Read this section of "Frequently
Asked Questions" on this web page.
Pose your questions to other
POPGENE users using the "USER's Forum" web page (Forum).
Why doesn't
POPGENE run on my computer ?
There are basically three reasons:
(1) You do not have the correct
hardware and software configuration. Please check the hardware and software
requirements for POPGENE in the "Quick User Guide".
(2) There is conflict between
POPGENE and other software running concurrently on your computer.
You get an error message " This program has performed an illegal
operation and will be shut down ". This has nothing to do with
our programming codes. It's between Borland C++, POPGENE's compiler
and other softwares. At this time, we know that Microsoft Plus and
pcANYWHERE32 are in conflict with POPGENE.
a. Microsoft Plus
You must install "Windows
95 Service Pack 1 Update". You can download this Windows 95 patch
from Microsoft via the internet at
b. pcANYWHERE32
POPGENE will not run
if you have pcANYWHERE32 in the background. This applies only to
the pcANYWHERE32 versions prior to version 7.5 (for Windows 95 AND
NT). You must exit pcANYWHERE32 in order to run POPGENE.
(3) There is a problem in what
you specified in the header section and/or your data in your data input
file. For example, you specified "7" loci but gave names for
six or eight loci, or that your data structure did not correspond to what
you specified. You will see a specific error message such as " The
population 1 individual 1 has incorrect number of loci ".
Why the "Help"
function only has Window related topics?
We have yet to completed the
"Help" function for POPGENE topics. We released POPGENE
without it because it has been our experience that users can run POPGENE
very effectively with assistance from the accompanying "Quick User
Guide".
We wish to state clearly that
our "Help" function will focus only on helping the users to run
the program. We do not have the financial means to develop a very detailed
"Help" function that will also guide you on population genetics
and statistics. We have listed all the relevant references in our "Quick
User Guide" and in the output. It is our hope that POPGENE
users are familiar with the genetic statistics that are available in POPGENE.
Who is
using POPGENE ?
POPGENE users are dedicated
plant and animal population geneticists worldwide, working in governments,
research organizations, and academic institutions.
What
is the process for upgrading POPGENE ?
All registered POPGENE
users will be informed by email of recent upgrades so that they can download
the latest version. This is the reason why you should register.
If you do not have email, use
the email address of a friend. We will only contact POPGENE users
by email due to cost consideration.
What are those files with
the "dat" extension in my POPGENE directory?
They are sample data files
for your convenience. Files "Diptest.dat" and "Haptest.dat"
are respectively, diploid and haploid test data for co-dominant markers.
Files "Rapddip.dat" and "Rapdhap.dat" are respectively,
diploid and haploid test data for dominant markers.
What is Shanon Index ?
Q. I am working with
RAPDs and I would like to use Shanon index because I have no information
concerning HWE. I would like to know precisely the formula used. Do you use
shanon index per locus or do you use it per primer ?
Answer. Use of Shannon index should be considered as the
last resort because it has little genetic interpretation except for measuring
the level of diversity. If the organism you work with is outcrossing species,
then the assumption of WHE may not be terribly wrong. It seems common in the
literature to make such assumption to estimate marker and null frequencies and
subsequently estimate population genetic structure (e.g., Lynch and Milligan,
1994. Mol. Ecol. 3:91-99; Chong et al. 1994. Curr. Genet. 26: 374-376). If the
organism is a selfing or inbred species, it is probably reasonably assumed that
the two RAPD phenotypes observed are likely to be marker and null homozygotes
because there is little heterozygosity particularly under selfing. In this you
can analyze the data as if it is from haploid individuals. It is more
appropriate to use Shanon Index for each locus than for each primer because each
RAPD band (locus) represents a different location over the genome where a DNA
sequence is complemnetary to the same primer. In this case, the formula for
Shannon's index is: -sum pi log2 pi, where pi is the frequency of the presence
or absence of the band (i=1, 2). The mean diversity may be estimated by an
average of index values over individual loci.
What are "Variable
as column" and "Record as column" under "Data Format"
in POPGENE command?
When the genotypes or haploidtypes
of each individual are recorded in columns on a line, as in the sample
data files, users should click on "Variable as column". However,
when the columns on a line are the genotypes/haploidtype at a single locus
for all individuals, users should click on "Record as column".
What are "Single populations",
"Groups" and "Multiple populations" under "Hierarchical
analysis" in POPGENE command?
When you analyse your data
for the first time, you should specify both "Single populations"
and "Multiple populations" so that you get the complete analysis
for your data.
POPGENE is interactive,
in the context that you can tell which populations and loci to include
in the analysis. Later on, you might just want to re-examine genetic parameters
of specific populations, individually, using just the "Single populations"
command. This will cut down on the amount of output for the analysis. Notice
that if you only specify "Single populations", you have no output
for those test statistics involving more than one populations (i.e.,
homogenetity test, genetic distance, dendrogram, gene flow, F-test) even
if you ask for them in the command boxes. If you want these test statistics,
you must also specify the "Multiple populations" command.
When you specify only the "Multiple
populations" command, you will have all those test statistics involving
more than one populations (i.e., homogenetity test, genetic distance,
dendrogram, gene flow, F-test), plus the remaining test statistic (i.e.,
allele frequency and number, Shannon index, HW-tests etc.) computed from
the overall data.
The "Groups" command
allows users to specify nesting of populations. Current version (1.21)
gives only 2 levels, for example, among species, among populations within
species, and within populations. We can program POPGENE for n-levels
of nesting. However, the problem is that the interactive part (i.e.,
when user has to specify the nesting of populations during the analysis
stage) becomes quite messy when users have many populations. We are still
working on the design logic.
We strongly recommend that
users try to run POPGENE using our sample data files. Specify the
different commands and examine the outputs. You will quickly master POPGENE.
What is the on-screen error
message in 16-bit version of POPGENE when I analyzed a large number of
populations, loci and alleles or when I have opened many files? Is there
a problem with the programming?
The answer is "NO".
There is no problem with the programming. Let us explain how POPGENE handles
the output of your data analyses.
POPGENE displays the results
of your data analyses on the computer screen. When you have many populations,
loci and alleles, and when you wanted to compute all available genetic
parameters, you would have hundreds/thousands of lines of output. When
the number of lines of output exceeded the limit of the POPGENE editor,
POPGENE could not display all your results on screen and you would receive
an error message. But don't worry, your results have already been saved
on your computer! This is because POPGENE automatically saved the complete
results of your data analyses into a file with the "rst" extension.
For example, if you specified that your data file was "My.dat",
then the complete results of your data analysis would be in file "My.rst".
So, open this file with your favorite word-processor and view the complete
results of your data analyses.
How much free RAM in your computer
would determine how many lines POPGENE could display on your screen. POPGENE
editor is RAM based to allow fast on-screen scrolling. So, try to free
up your computer RAM when running POPGENE by closing other programs and
also screens that are not in use within POPGENE. For example, you might
be running several data files with POPGENE. Try to close those on-screen
data and output files before you call up a new data file for analyses.
The editor for the 32-bit version
of POPGENE does not have such problem unless the computer only has 8 MB
RAM and limited free disk space due to advanced memory management under
Windows 95 & NT.
When
is the module for quantitative traits available in POPGENE ?
We want to beta test it later
this year and distribute it in 1998. Right now, the module under development
can compute a number of parameters, including Fst
for individual quantitative traits. We still have to incorporate the estimation
of confidence intervals based on bootstrapping, the comparison between
Fis from marker genes and quantitative traits,
and the among-population comparison of multitrait covariance matrices (phenotypic/genetic/environmental).
You can read some of these in a paper by Yang, Yeh and Yanchuck in Genetics
142:1045-1052 (1996).
Notice that the analysis of
quantitative traits will be computationally demanding. You will need a
fast CPU (Pentium-based PC) and much more random access memory (RAM) than
for analyzing co-dominant and dominant markers.
When will you release a
POPGENE version for other platforms, such as Mac OS and Unix ?
We do have a large number of
inquiries concerning POPGENE for the Mac OS, especially from Europe.
At this time, we have no plan to develop POPGENE for other computer
platforms. This is strictly an economic decision because we must secure
additional fund to finance this work. POPGENE has over 60,000 lines of
codes, and many built-in library functions which are specific to the Borland
C++ compiler that we used. The compiled codes exceeded 3,490,000 lines
and it will be naive to think that Mac conversion will be an easy task.
However, Mac users can run
POPGENE on PowerPc, but must first install a software such as "Virtual
PC" from Connectix on the PowerPc. "Virtual PC" uses between
150-260MB of your disk as if it is a "C" drive (in PC terminology)
and installs Windows 95 on it. The 32-bit version of POPGENE rans
effortlessly. To be able to use "Virtual PC", your PowerPc processor
must be the 604 or 604e chip running at speed of 100Mhz and higher; or
any 603e chip running at 180Mhz or faster. The PowerPc must also have Mac
OS 7.5.5 or later and with a minimum of 24 MB RAM.
POPGENE should also run under
"Soft Windows" on PowerPc, but we have not tested POPGENE under
"Soft Windows". We would appreciate those with "Soft Windows"
to test run POPGENE and tell us the outcome.
Send mail to francis
yeh with questions or comments about this web site.
Last modified: 十月 04, 2001