No Hits

[bibilujan]

Hi,

I am interested in running this tool to possibly visualize and pull out some symbionts from my sample. I was able to use bloobtools to and get the graph and table with view and blobplot commands, but I get all my entries as "no hits". Would you be able to help me to troubleshoot this issue? My assembly is of a 260MB plant, closely related to A. thaliana so there should be lots of hits that are properly annotated.

Here is the summary stats:

C.microcarpa-plot-4.C.microcarpa.BlobDB.json.span.phylum.p7.100.bestsum - spades Group colour count visible (%) span visible(%) n50 GC GC (std) cov_mean cov_std read map read map (%) all None 49,622 100.0% 203,388,998 100.0% 107,031 0.41 0.1 25.2 172.8 0 0.0% no-hit #d3d3d3 49,622 100.0% 203,388,998 100.0% 107,031 0.41 0.1 25.2 172.8 0 0.0%

And a few lines from the "view" output (the last is a hit to A. thaliana):

NODE_1_length_1111123_cov_10.2151_ID_114688196 6429 gi|727511611|ref|XM_010433800.1| 97.47 3795 44 29 708277 712041 3773 1 0.0 NODE_1_length_1111123_cov_10.2151_ID_114688196 5991 gi|727522685|ref|XM_010439039.1| 99.57 3286 14 0 992226 995511 6183903 0.0 NODE_1_length_1111123_cov_10.2151_ID_114688196 5485 gi|727547328|ref|XM_010448413.1| 95.46 3483 80 44 708279 711744 3425 4 0.0 NODE_1_length_1111123_cov_10.2151_ID_114688196 4427 gi|7270623|emb|AL161590.2| 87.72 3958 280 121 708239 712136 166312 170123 0.0

[DRL]

Hi bibilujan,

your blast file has to be in the format '6 qseqid staxids bitscore ...', while your blast results seem to be in the format '6 qseqid staxids subject ...'

I recommend re-running the blast with :

-outfmt '6 qseqid staxids bitscore std sscinames sskingdoms stitle'

that should fix the issue

cheers,

dom

[bibilujan]

Hi Dom,

The results you see I actually got after running the blast in the way you are specifying, here is the command:

blastn -task megablast -query ../Assembly_C.microcarpa/Old_Assemblies/C.microcarpa-m13c-filtered-200.fasta -db nt-ref-db/nt -outfmt '6 qseqid staxids bitscore std sscinames sskingdoms stitle' -max_target_seqs 25 -culling_limit 2 -num_threads 16 -evalue 1e-25 -out C.microcarpa13.vs.nt.25cul1.1e25.megablast.out

I deleted the third row from my blast results and re-ran blobplot, and I did get hits this time, however im not sure why this is happening or if I can trust these results. What do you think?

Thank you for taking the time to look at this,

Beatriz

C.microcarpa-plot.C.microcarpa.BlobDB.json.span.phylum.p7.100.bestsum.stats.txt

[DRL]

Hi Beatriz,

nice blobplot! :)

several things:

• the blobplot looks good, in the sense that blobtools could parse taxonomy information from the blast result(s). So in theory everything should be good, although I'd like to check the blastfile myself.
• could you tar -zcvf your blast result(s) and sent me a dropbox link to it so I can check if there's something wrong there?

Regarding the plot:

• you have lots of proteobacteria in there (which is fine)
• you have ~50 Mb of "undef". That is a result from blast hits to sequences that do not have the taxonomic rank of "phylum" assigned to them (according to NCBI taxonomy). That happens sometimes. You want to check what taxids are causing this and maybe run the blobplot command again on another rank (e.g., order). Maybe these are hitting your target organism, but return "undef" because your target organism might not have a proper taxonomy (in the NCBI taxonomy/nt database you are using)
• You might want to check when your taxonomy/nt database has last been updated

hope this helps,

dom

[bibilujan]

Here you can find the blast results, this is before removing the third row:

https://www.dropbox.com/s/l9iwvc3i9wdjulb/C.microcarpa_blast_results.tar?dl=0

Is it common to have so much proteobacteria? I was hoping to find some fungi in there. I did re-download the most recent nt database and taxonomy, because I thought that was why I was not getting any hits initially.

Thanks,

Beatriz

[bibilujan]

I had placed a damaged file on dropbox, but I have replaced it with the correct one now.

https://www.dropbox.com/s/l9iwvc3i9wdjulb/C.microcarpa_blast_results.tar?dl=0

Thanks,

Beatriz

[DRL]

Hi Beatriz,

the problem with the blast file is that the second column does not carry the taxid:

• your format is `qseqid bitscore std ...``

This could be because you are missing the taxonomy database from NCBI and therefor BLAST can not assign a taxid to the hits. Another possibility is that the taxdb database is not in the path defined by the BLASTDB environment variable.

cheers,

dom