Hello QIIME2 team,
I am running Deblur with the following command using QIIME2 (2024.5):
qiime deblur denoise-other
--i-demultiplexed-seqs reads_qza/reads_trimmed_joined.qza
--i-reference-seqs ref_seq.qza
--p-trim-length 385
--p-min-reads 1
--p-min-size 1
--p-sample-stats
--p-jobs-to-start 48
--output-dir deblur_output_nofilter
I am seeing a much larger read loss than expected—sometimes below 25%.
I have already set both --p-min-reads 1 and --p-min-size 1 to keep as many features as possible, and am using a fairly low (385) trim length despite a very uniform length distribution.
-
Is there any way to further pinpoint why Deblur is discarding these reads?
-
Do you recommend adjusting other parameters or performing additional pre-filtering to retain more reads while using Deblur?
-
Could the reference sequence (ref_seq.qza) filtering play a role, even though it is permissive?
-
Are there log files or statistics I should inspect to better diagnose where the reads are lost (e.g., positive filter, chimera, quality filter, etc.)?
Attached are my sequence length summary and specific read loss counts per sample for reference.
Thank you for your help!
| Column 1 | Column 2 | Column 3 | Column 4 |
|---|---|---|---|
| Sample_ID | Merged_reads_count | Read no after filter with deblur | |
| 16s_1 | 30089 | 6474 | |
| 16s_2 | 53816 | 12645 | |
| 16s_3 | 79655 | 27131 | |
| 16s_4 | 79845 | 37235 | |
| 16s_5 | 79858 | 26578 | |
| 16s_6 | 80235 | 33970 | |
| 16s_7 | 79779 | 31072 | |
| 16s_8 | 80168 | 29701 | |
| 16s_9 | 79607 | 35456 | |
| 16s_10 | 79770 | 35650 | |
| 16s_11 | 79983 | 29191 | |
| 16s_12 | 79930 | 43857 | |
| 16s_13 | 79747 | 36896 | |
| 16s_14 | 79988 | 32404 | |
| 16s_15 | 79924 | 33262 | |
| 16s_16 | 79798 | 24406 | |
| 16s_17 | 79845 | 37163 | |
| 16s_18 | 79972 | 37212 | |
| 16s_19 | 30172 | 10326 | |
| 16s_20 | 32868 | 7742 | |
| 16s_21 | 80091 | 41318 | |
| 16s_22 | 79779 | 34413 | |
| 16s_23 | 79951 | 29166 | |
| 16s_24 | 80160 | 30924 | |
| 16s_25 | 79882 | 33057 | |
| 16s_26 | 79560 | 31351 | |
| 16s_27 | 79968 | 30205 | |
| 16s_28 | 79577 | 19150 | |
| 16s_29 | 79943 | 33613 | |
| 16s_30 | 79996 | 37147 | |
| 16s_31 | 26324 | 6729 | |
| 16s_32 | 29810 | 6486 |
