Easier to use randomisation results

This is an updated version of a previous post.  The key difference is that the significance classification described previously was too confusing, as values could be positive or negative and became more significant as they approached zero.  Instead, Biodiverse now provides relative ranks which can easily be converted to significant/not significant for any alpha cutoff.  This change should not affect many users, as the 0.99_003 release containing it was never announced...

One of the issues users face with the randomisations in Biodiverse is what to do with them once they have been run.

A key point is that the results are stored on the other analysis objects themselves, as extra indices and lists.  The index names themselves are a bit cryptic, but are consistent, and there is a description of what they mean here:  https://purl.org/biodiverse/wiki/AnalysisTypes#where-do-the-randomisation-results-go-and-what-do-they-mean

Even then, it can be difficult working out which of your groups have index scores that are significantly different from the set of randomised results.  This is because the data are plotted as a continuum (which in turn is because it uses the same plotting process as the original index scores).  The first image below is an example of this plotting.

One can easily export the data and work with them in a GIS or stats package, but any tied values need to be factored in for lower tail tests, for example as used in the CANAPE process.

With the next version of Biodiverse this categorisation will become a little easier.  Biodiverse will automatically calculate rank-relative positions that can be easily converted into significance levels.  These are stored in new lists that can be displayed and exported in the same way as any other data.

As an example, imagine you have run a randomisation analysis for a BaseData containing a spatial analysis in which you calculated phylogenetic endemism.  Assume that the randomisation's name is rr (not a good name, but it's convenient to type here), so the spatial analysis will now have three lists you can plot.  The first two are the same as ever:  SPATIAL_RESULTS contains the observed results for each group (cell), and rr>>SPATIAL_RESULTS contains indices to track the randomisations for each index in SPATIAL_RESULTS.  For example, for PE_WE there will be C_PE_WE, Q_PE_WE, T_PE_WE and P_PE_WE collating, respectively, the number of times observed PE_WE was higher than that generated using the randomised data, the number of times observed PE_WE was compared against the scores from the randomised data, the number of times the observed and random scores were tied, and the proportion of iterations that the observed score was higher than the random scores (P_PE_WE = C_WE_PE / Q_PE_WE).

The new list is rr>>p_rank>>SPATIAL_RESULTS.  This contains a set of results using the same names as the original indices in SPATIAL_RESULTS, but converted to their rank relative positions.  Importantly, the lower tail ranks take into account any ties in the comparisons, thus simplifying any code that uses theses results.  Also, any value that would be considered not significant at alpha=0.05 (one tailed, high or low) is converted to undef (null).  This makes any plots of the results clearer within Biodiverse so one can more easily see which groups would pass a one-tailed high or low test.

An example plot is in the second image below.

All the ranks have been combined into the same list to reduce the number of indices and lists generated, and thus use less memory and disk space (an index cannot be simultaneously significantly high and low so there is no overlap).  If you are interested in a one tailed test for high values then ignore the low values, and vice-versa.  The values can be easily separated after exporting the results.

Currently the results are plotted in the same manner as any data, but there are plans to allow users to overlay the randomisation significance results over the top of the observed results, for example by masking our any non-significant scores for a given threshold.

The current system plots all scores, regardless of whether they pass a threshold or not.  This is useful, but is difficult to interpret when looking for significance against the randomisation.

The new randomisation list contains indices for the rank-relative positions of the observed values against the randomly generated values. These can then be used for one and two tailed significance tests.  The plotting could be improved, e.g. in this case it appears there are only two values, but this is simply due to the colour scaling.  However, it works well enough now for exploration - proper maps can also be made using a GIS or stats package.

To try it out you will need the 1.99_004 release (or later).  It can be accessed from https://github.com/shawnlaffan/biodiverse/wiki/Downloads

This is a new implementation, so any feedback about usability would be very useful.  

Shawn Laffan

29-Aug-2016

For more details about Biodiverse, see http://purl.org/biodiverse 

For the full list of changes in the 1.99 series (leading to version 2) see https://purl.org/biodiverse/wiki/ReleaseNotes 

To see what else Biodiverse has been used for, see https://purl.org/biodiverse/wiki/PublicationsList

You can also join the Biodiverse-users mailing list at http://groups.google.com/group/Biodiverse-users 

New, more efficient file format

Users of Biodiverse will perhaps be familiar with what is called the "native" format for basedata, trees, matrices and projects.  These are the .bds, .bts, .bms and .bps files that are created when you save these objects.

The reality is that the "native" format is just a serialisation format in which all the various parts of the perl data structures that make up an object (e.g. a tree) are converted to a format that can be written to disk and then re-read at a later date, possibly on another computer.

While the format we have been using (called Storable) is stable and has done a good job over the years, a newer, more efficient format called Sereal is now available.  Version 2 of Biodiverse will use this new format by default.

The main reason for shifting to the Sereal format is efficiency: saving files is faster, and the file sizes are smaller.  See details here: http://blog.booking.com/sereal-a-binary-data-serialization-format.html 

These size and speed improvements will not be very noticeable for small files, but it can all add up when one is working with tens of thousands of groups (e.g. cells) and thousands of labels (e.g. species) across hundreds of spatial and cluster analyses.  A quick experiment with such a data set resulted in a greatly reduced file size (~1.6GB to ~750MB), with the time taken to save to file reducing from 30s to 12s.  The file load times were about the same at ~20s.  (Admittedly this was not a very scientific experiment, but the results were consistent across multiple runs).

What do users need to be aware of?  The main thing is that files created in Biodiverse version 2 will not be backwards compatible.  This means that Biodiverse version 1.1 or earlier will not be able to open files created using version 2 by default.  However, the "save as" dialogues have the option to save to the old format so you can maintain compatibility with older versions if you are in a mixed environment.

Also, any file in the old format that is loaded into Biodiverse version 2 will still be saved using the old format unless the user explicitly saves it to the new format.

If you want to test the new file format then it will be available in the 1.99_004 development release which should be coming out within the next week.

Shawn Laffan

29-Aug-2016

For more details about Biodiverse, see http://purl.org/biodiverse 

For the full list of changes in the 1.99 series (leading to version 2) see https://purl.org/biodiverse/wiki/ReleaseNotes 

To see what else Biodiverse has been used for, see https://purl.org/biodiverse/wiki/PublicationsList

You can also join the Biodiverse-users mailing list at http://groups.google.com/group/Biodiverse-users 

Biodiverse now categorises your randomisation results

[[ 2016-08-30:  This post has been superseded - the categorisation was not sufficiently clear.  See this followup post for how the system works now ]]


One of the issues users face with the randomisations in Biodiverse is what to do with them once they have been run.

One key point is that the results are stored on the other analysis objects themselves, as extra indices and lists.  The index names themselves are a bit cryptic, but are consistent, and there is a description of what they mean here:  https://purl.org/biodiverse/wiki/AnalysisTypes#where-do-the-randomisation-results-go-and-what-do-they-mean

Even then, it can be difficult working out which of your groups have index scores that are significantly different from the randomisation.  This is because the data are plotted as a continuum (which in turn is because it uses the same plotting process as the original index scores).  The first image below is an example of this plotting.

One can easily  export the data and work with them in a GIS or stats package, but any tied values need to be factored in for lower tail tests, for example as used in the CANAPE process.

With the next version of Biodiverse this categorisation will become a little easier.  Biodiverse will automatically categorise your randomisation results into significance levels, putting the results into new lists on the objects that can be displayed and exported in the same way as any other data.

As an example, imagine you have run a randomisation analysis for a BaseData containing a spatial analysis in which you calculated phylogenetic endemism.  Assume that the randomisation's name is rr (not a good name, but it's convenient to type here), so the spatial analysis will now have three lists you can plot.  The first two are the same as ever:  SPATIAL_RESULTS contains the observed results for each group (cell), and rr>>SPATIAL_RESULTS contains indices to track the randomisations for each index in SPATIAL_RESULTS.  For example, for PE_WE there will be C_PE_WE, Q_PE_WE, T_PE_WE and P_PE_WE collating, respectively, the number of times observed PE_WE was higher than that generated using the randomised data, the number of time observed PE_WE was compared against the scores from the randomised data, the number of times the observed and random scores were tied, and the proportion of iterations that the observed score was higher than the random scores (P_PE_WE = C_WE_PE / Q_PE_WE).

The new list is rr>>sig>>SPATIAL_RESULTS.  This contains a set of categorisations for one and two tailed tests for each index found in SPATIAL_RESULTS.  The lower tail tests take into account any ties in the comparisons.  An example plot is in the second image below.

For the PE_WE example, one has SIG_1TAIL_PE_WE and SIG_2TAIL_PE_WE.  SIG_1TAIL_PE_WE is a one-tailed test for higher or lower than expected.  It has a value of 0.01 if it is significantly high at alpha=0.01, 0.05 if high for alpha=0.05, -0.01 if it is significantly low at alpha=0.01, and -0.05 for low at alpha=0.05.  If it is not significant then it has a null (undefined) value.

SIG_2TAIL_PE_WE has the same numbers, but for a two tailed test.  Values of -0.05 and 0.05 are low or high for a two tailed alpha=0.05, i.e. the observed scores are in the outer 5% of the distribution of random scores (<2.5% or >97.5%, respectively), while those with -0.01 or 0.01 are in the outer 1% and significant at alpha=0.01.

The upper and lower one-tailed tests have been combined into the same list to reduce the number of indices and lists generated, and thus use less memory and disk space (an index cannot be both significantly high and low so there is no overlap).  If you are interested in a one tailed test for high values then ignore the low values, and vice-versa.  The values can be easily separated after exporting the results.

Currently the results are plotted in the same manner as any data, but there are plans to allow users to overlay the randomisation significance results over the top of the observed results, for example by masking our any non-significant scores.

The current system plots all scores, regardless of whether they pass a threshold or not.  This is useful, but is difficult to interpret when looking for significance against the randomisation.

The new categorisation filters out any score that is not significant at an alpha level of either 0.05 or 0.01 (here the one-tailed results are plotted, so negative values are significantly low).  The plotting could be improved, but it will work well enough now for exploration - proper maps can also be made using a GIS or stats package.  

This is a new implementation, so any feedback about usability would be very useful.  

Shawn Laffan

06-Aug-2016

For more details about Biodiverse, see http://purl.org/biodiverse 

For the full list of changes in the 1.99 series (leading to version 2) see https://purl.org/biodiverse/wiki/ReleaseNotes 

To see what else Biodiverse has been used for, see https://purl.org/biodiverse/wiki/PublicationsList

You can also join the Biodiverse-users mailing list at http://groups.google.com/group/Biodiverse-users