i'm do the pipeline of ASV for 16s analysis with demultiplexed files with .fq format.

> list.files(path)
 [1] "DAtos demulteplixados.Rproj"         "filtered"                           
 [3] "Mock_S280_L001_R1_001.fastq"         "Mock_S280_L001_R2_001.fastq"        
 [5] "SAMPLE1_1.fq"                        "SAMPLE10_1.fq"                      
 [7] "SAMPLE11_1.fq"                       "SAMPLE12_1.fq"                      
 [9] "SAMPLE13_1.fq"                       "SAMPLE14_1.fq"                      
[11] "SAMPLE15_1.fq"                       "SAMPLE2_1.fq"                       
[13] "SAMPLE3_1.fq"                        "SAMPLE4_1.fq"                       
[15] "SAMPLE5_1.fq"                        "SAMPLE6_1.fq"                       
[17] "SAMPLE7_1.fq"                        "SAMPLE8_1.fq"                       
[19] "SAMPLE9_1.fq"                        "silva_nr_v128_train_set.fa.gz"      
[21] "silva_species_assignment_v128.fa.gz"
>fnFs <- sort(list.files(path, pattern="_1.fq", full.names = TRUE))
> sample.names <- sapply(strsplit(basename(fnFs), "_"), `[`, 1)

> sample.names
 [1] "SAMPLE1"  "SAMPLE10" "SAMPLE11" "SAMPLE12" "SAMPLE13" "SAMPLE14" "SAMPLE15" "SAMPLE2" 
 [9] "SAMPLE3"  "SAMPLE4"  "SAMPLE5"  "SAMPLE6"  "SAMPLE7"  "SAMPLE8"  "SAMPLE9" 
> plotQualityProfile(fnFs)
Error: record does not start with '@'

Help my please.

This error comes from the ShortRead package, and it means that your FASTQ files have non-standard names (which should start with an @). As an example (from the command line):

$ zcat 303422_S150_L002_R1_001.fastq.gz | head -n 4 
@A00177:32:HFVCFDMXX:2:1101:1108:1000 1:N:0:GTCAACAG+GTCAACAG <- Note that this is a correct seqname
NGGATGGATCACTGGGTGTGTGGCATAGTCCATGCTCTTAACCAGATTTGAACAGGAGAATGGCCACCTGGCCCAGGTAGAAGTAGATGAAGTGTTTGGTT
+
#FFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFF:FFFFFFFFFFF

## now make a FASTQ with missing @

$ zcat 303422_S150_L002_R1_001.fastq.gz | sed  's/@//g'  | head -n 24 > tmp.fq
$ head -n 4 tmp.fq 
A00177:32:HFVCFDMXX:2:1101:1108:1000 1:N:0:GTCAACAG+GTCAACAG <- this is not!
NGGATGGATCACTGGGTGTGTGGCATAGTCCATGCTCTTAACCAGATTTGAACAGGAGAATGGCCACCTGGCCCAGGTAGAAGTAGATGAAGTGTTTGGTT
+
#FFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFF:FFFFFFFFFFF


$ Rscript -e "suppressPackageStartupMessages(library(ShortRead)); qa('tmp.fq')"
Error: record does not start with '@'
Execution halted

So you need to figure out what the problems is with your FASTQ files.