Error running q2-picrust2 in both full-pipeline and custom-tree-pipeline

Dear everyone,

I have a weird issue with picrust2, running it as QIIME plugin. I tried two approaches (see below A and B ) and both result in errors that I cannot explain and that I have not seen others reporting.
In my dataset there are about 96 samples and I expect about 3000 ASVs. I am running on Ubuntu system, 16 cores and 32 GBs memory.
It does not seam to be memory related problem, which I saw many users experienced. And approach A I that I tried below, should help in this regards.

I would be very glad for anyone who could help me solve this problem.

qiime picrust2 custom-tree-pipeline --i-table table.qza \
                                    --i-tree rooted-tree.qza \
                                    --output-dir q2-picrust2_output_tree \
                                    --p-threads 14 \
                                    --p-hsp-method pic \
                                    --p-max-nsti 2 \
                                    --verbose

And the error output is:

Running the below commands:
hsp.py -i 16S -t /mnt/scratch/tmp/tmpqpvs06tz/placed_seqs.tre -p 1 -n -o /mnt/scratch/tmp/tmpqpvs06tz/picrust2_out/16S_predicted.tsv.gz -m pic -e 0.5

Error running this command:
hsp.py -i 16S -t /mnt/scratch/tmp/tmpqpvs06tz/placed_seqs.tre -p 1 -n -o /mnt/scratch/tmp/tmpqpvs06tz/picrust2_out/16S_predicted.tsv.gz -m pic -e 0.5

Standard error of the above failed command:

Error running this command:
Rscript /home/mattesa/miniconda3/envs/qiime2-2021.11/lib/python3.8/site-packages/picrust2/Rscripts/castor_hsp.R /mnt/scratch/tmp/tmpqpvs06tz/placed_seqs.tre /mnt/scratch/tmp/tmp03sy3041/subset_tab_0 pic 0.5 FALSE FALSE /mnt/scratch/tmp/tmpygqyute4/predicted_counts.txt /mnt/scratch/tmp/tmpygqyute4/predicted_ci.txt 100

Standard error of the above failed command:
Error: None of the reference ids within the function abundance table are found within the input tree. This can occur when malformed or mismatched custom reference files are used.
Execution halted
qiime picrust2 full-pipeline --i-table table.qza \
                             --i-seq rep-seqs.qza \
                             --output-dir q2-picrust2_output_full \
                             --p-placement-tool sepp \
                             --p-threads 14 \
                             --p-hsp-method mp \  
                             --p-max-nsti 2 \
                             --verbose

the error message is as follow

Warning - 2 input sequences aligned poorly to reference sequences (--min_align option specified a minimum proportion of 0.8 aligning to reference sequences). These input sequences will not be placed and will be excluded from downstream steps.

This is the set of poorly aligned input sequences to be excluded: dd2cfc63a7cc6581f5b03a043994cf55, 7b8376e5d18ae5cb3ad1df4a57d7be64



Error running this command:
hsp.py --tree /mnt/scratch/tmp/tmp79ljaac9/picrust2_out/out.tre --output /mnt/scratch/tmp/tmp79ljaac9/picrust2_out/marker_predicted_and_nsti.tsv.gz --observed_trait_table /home/mattesa/miniconda3/envs/qiime2-2021.11/lib/python3.8/site-packages/picrust2/default_files/prokaryotic/16S.txt.gz --hsp_method mp --edge_exponent 0.5 --seed 100 --calculate_NSTI --processes 1

Standard error of the above failed command:

Error running this command:
Rscript /home/mattesa/miniconda3/envs/qiime2-2021.11/lib/python3.8/site-packages/picrust2/Rscripts/castor_hsp.R /mnt/scratch/tmp/tmp79ljaac9/picrust2_out/out.tre /mnt/scratch/tmp/tmptsv16bdb/subset_tab_0 mp 0.5 FALSE FALSE /mnt/scratch/tmp/tmp4d6ir34n/predicted_counts.txt /mnt/scratch/tmp/tmp4d6ir34n/predicted_ci.txt 100

Standard error of the above failed command:
Error:
Error - at least one trait in the prediction table was entirely missing values.
Execution halted