dcdanko / capalyzer Goto Github PK

Rarefied Chao1 and raw give very different results

> for n in [100, 1000, 10 * 1000, 100 * 1000, 1000 * 1000, 0]:
    print(table_factory.taxa_alpha_diversity(metric='chao1', rarefy=n)['SL280226'])
95.37651234634326
253.86701478272045
493.5870200957671
650.3338915219408
701.496205789874
2182.7178442962995

The measured richness (raw, not Chao1) of the unrarefied sample is 704. The sample has 1,052,098 taxonomic reads.

Should be a core part of taxa parsing https://github.com/dcdanko/capalyzer/blob/master/scripts/modified_kraken_parser.py

Hi,
I am interested in performing the diversity analysis using the Metasub data.
I first installed capalyzer in python 3.7 and then I copy and paste the commands below in a script and ran it:

python3 diversity_estimatives.py

file (diversity_estimatives.py) content:
$from capalyzer.packet_parser import DataTableFactory

$table_factory = DataTableFactory("/home/users/luiza.araujo/Documents/CIPE/NGS/Microbiome/MetaSub/Jan_2019/sao_paulo")

$krakenhll_richness = table_factory.taxa_alpha_diversity(metric='richness', rarefy=1000000) # $krakenhll is the default tool
$metaphlan2_entropy = table_factory.taxa_alpha_diversity(tool='metaphlan2') # entropy is the default metric

It ran without any errors, however I cant find the output files with the diversities.
Can you help me with that?

Thanks!

PS: the /sao_paulo folder was downloaded from the /city/packets dropbox link.

Hi,
Could you help me with the following questions?

1- could you let us know the main differences between krakenhll and metaphlan?
2- Does Capalizer also calculates beta diversity (Bray-curtis, Jaccard and weighted/unweighted unifrac distances)?
3- In the diversities estimation, how could we use specific taxonomic levels (ex: analyzed by species or gender...)

Thanks!

Relates to Issue #6 and #7

We should support diversity calculations at any rank, however this requires data packets to include multi-rank taxa tables which is still in development.