blastx diamond returns taxid "NA" for some sequence queries

Question

[pedrorodrigues]

Hello, I ran blastx against the nr database, using diamond (ran at a hpc cluster). After analysing the results, I noticed that some sequences returned no taxids, although other information such as pident or scientific names were returned. See below a sample of the results I obtained

columns are: qseqid evalue bitscore length pident stitle qcovhsp sscinames staxids

1 TRINITY… 6.20e- 92 283.    202  71.8 XP_031787202.1…  70.9 Nasonia vi… 7425  
 2 TRINITY… 1.20e-108 330.    164  99.4 NP_729590.1 un…  90.9 Drosophila… 7227;…

3 TRINITY… 3.70e- 25 108.     50 100   XP_020808035.1…  55.1 Drosophila… 7274  
 4 TRINITY… 3.20e- 48 168.    100  71   XP_014215226.1… 100   Copidosoma… 29053 

5 TRINITY… 5.10e-121 352.    172 100   HAH0498887.1 w… 100   N/A         NA    
 6 TRINITY… 1.30e- 11  67.4    46  71.7 EEY2123519.1 s…  54.3 N/A         NA    
 7 TRINITY… 2.10e-104 314.    158 100   ABC86463.1 IP0…  86   Drosophila… 7227  
 8 TRINITY… 5.20e- 19  86.3    45 100   WP_021218988.1…  55.1 Pseudomona… 43263…

Rows 5 and 6 show "NA" as a result for taxids. But when I search ncbi for the sequence name ("stitle"), I can find a taxid for both these sequences (in this case, E. coli, taxid 562). I have downloaded the ncbi taxdmp and setup the diamond database built to include taxonomy.

Have you ever had this problem, i.e. some sequences failing to receive a taxid during a blastx search? Thank you in advance for any help!

[GenoMax]

Looks like those hits are to accessions in Identical Protein Groups database. My guess is those are not represented in the NCBI taxonomy.