Soryy, I didn't explain my data clearly. Actually, my data was PGM sequencing data. I've merged the barcode&primer-removed sequence data (fasta) and qualitydata (.qual) into .fastq data like this:
I've read the post from How to do multiplex of Ion torrent sequences. I guess this solution could work for me, since I don't have barcode, either. but I am stuck in one problem: instead of split by samples, now all my samples are in one fastq file. I tried in qiime1 to split it with split_libraries_fastq.py, but it requires either "barcode read fastq files" or "sample_id ( --sample_ids:
Comma-separated list of samples ids to be applied to all sequences, must be one per input file path (used when data is not multiplexed))". I guess sample_id is not practical since typing ids manually...uh...scaring...
Also, I've merged the barcode&primer-remained sequence data (fasta) and qualitydata (.qual) into .fastq data like this:
but I really do not know what to do next 
or, I also tried to demux the data via qiime1, and get the output sequence file XXX.fna. Then I imported it in qiime2 as "--type FeatureData[Sequence]" and try to use dada2 to generate the "representative-sequences XXX.qza" "table xxx-dada2.qza" "denoising-stats xxx-dada2.qza", BUT it returned " Argument to parameter 'demultiplexed_seqs' is not a subtype of SampleData[PairedEndSequencesWithQuality | SequencesWithQuality]"
I am totally in a mess now.... could you pleas give me any suggestion to start?
Thanks a lot!
Sincerely,
Nan