Parsing MEGAN blast2lca output

Question

[qinglong]

Hi there,

I am seeking help from the community for parsing blast2lca output from MEGAN version 6:

blast2lca output (semi-colon separated; I have millions of genes in a txt file; 1001, 10010, 10011 etc. are gene ID):

1001; ;g__Bacteroides; 100;s__Bacteroides caccae; 21;
10010; ;g__Clostridium; 100;s__Clostridium butyricum; 50;
10011; ;g__Clostridium; 100;s__Clostridium butyricum; 75;
...
...

Here is what I want (set cut-off to 50; stored in a tab-delimited file):

GeneID  Genus           Species
1001    Bacteroides 
10010   Clostridium Clostridium butyricum
10011   Clostridium Clostridium butyricum
...
...

Could you please provide me a command or simple script to do this ?

Much appreciated!

Qinglong

[cpad0112]

if input and output are exactly same as OP, you can have this way:

test.txt is input and output is:

$ awk -v OFS="\t" -F '[;__]' 'NR==1 {print "GeneID", "Genus","Species"};{print $1,$5,$9}' test.txt       

GeneID  Genus   Species
1001    Bacteroides Bacteroides caccae
10010   Clostridium Clostridium butyricum
10011   Clostridium Clostridium butyricum

input:

$ cat test.txt 
1001; ;g__Bacteroides; 100;s__Bacteroides caccae; 21;
10010; ;g__Clostridium; 100;s__Clostridium butyricum; 50;
10011; ;g__Clostridium; 100;s__Clostridium butyricum; 75;

[qinglong]

Thanks!!! But I also need to have a filtering based on the confidence score (cut-off: 50), do you have any other command to do that?

[cpad0112]

 $ awk -v OFS="\t" -F '[;__]' 'NR==1 {print "GeneID", "Genus","Species"}; { if ($10<=50) print $1,$5,$9,$10}' test.txt  
GeneID  Genus   Species
1001    Bacteroides Bacteroides caccae   21
10010   Clostridium Clostridium butyricum    50