dcdanko / md2 Goto Github PK

Write a README that explains

• Project overview and goals
• Sources of data
• Data schema by kingdom
• etc.

• Starting with OTU tables, add names of the microbe and convert the abundances for each Studies

https://inport.nmfs.noaa.gov/inport/item/30877

Write a script that takes the output of the script in Issue #1 and does the following

• Removes all taxa below the rank of species
• Breaks the taxa table into virus, bacteria, fungi

Add some additional columns based on the MetaSUB taxa table

Write a set of directions for building the MD2 table from scratch

Counts for specific location (i.e. Tara Oceans are confusing) convert them to always sometimes never or similar

Write code to compare taxonomic profiles that include relative abundance information. Relative abundance info should be considered as the probability of selecting a particular taxa in a sample and as such can be appropriately measured using the Kolmogorov-Smirnov test.

This should not replace the comparisons that do not use abundance information.

I installed the Microbe directory 2 locally on my Linux system and want to achieve the following from it.
Give a phenotypic and ecologic description of microbial species parameters.
Annotate the microbiome where taxa have been identified.

I generated a test .csv that contains all the bacterial species from one of our samples. I am trying to give a genotypic phenotypic and ecologic description of microbial species parameters. How do I provide this file to microbe directory 2 and generate a file which contains all filled in details?

thanks for your help :)

Add a script that fills in taxonomic ranks from their relatives.

e.g.

• If a genus has n/a for spore forming check all its children and set the value to always/sometimes/never/na.
• If a species has n/a for gram-stain check if gram-stain is defined by one of its parents

Thank you for providing such a useful resource! Maybe I am missing something, but is the code that was used for extracting microbial attributes (such as gram-stain, spore-formation, metabolic attributes, etc ...), ie the regular expressions that were used to automatically extract certain attributes from the mentioned databases / studies, available?

Write a script that turns the NCBI taxonomy into a table with three columns

• Taxa ID
• Scientific Name
• Taxonomic Rank

Take two files with MetaPhlan/Kraken data and compare them!

• Try Chi-square test on the two sample data!

Write a script that matches the output from Issue #2 to our existing tables from Microbe Directory 1.0 and. other sources.

Follows Issue #12

We can't fill in every single empty value in MD2. We need a plan to triage which columns and which rows are the most useful and tractable. This will inform how the survey is built.

Basic criteria:

• empty columns are more important
• draft species are less important
• common species are more important

Follows Issue #3

Write a script that cleans the table output by Issue #3. Should do the following

• rename certain columns with confusing names (i.e. 'microbiome_location')
• normalize column name format to all lowercase, alphanumeric and underscore only (snakelowercase)
• remove taxonomy (for npw)
• merge list columns (refseq-id, etc) into comma separated lists
• merge duplicate rows
• reomve citation columns

Generate some statistics for each column and some summary statistics over riws so we can evaluate which rows need the most input