synonymous/non-syn. SNPs refer to protein coding regions. The latter alters the aminoacid while the first does not. So first of all you should be sure that the genomic sequences above are in exonic regions. Apart from that, a SNP is defined as such when it happens with a certain frequency in a population (not sure about the threshold though).

btw, I don't understand the dashes in your sequence. Does this represent an alignment?

Simply comparing two sequences of unknown quality makes it difficult to identify SNPs. If the 2 seqs are single reads from a high-throughput device, I would be hesitant to call any polymorphisms. If the above are high-quality and/or aligned to a reference genome, then you can begin to call SNPs and map them to a genome and its encoded genes.

So, with what you present above, you can perform a translated BLAST search (BLASTX) with seq as query against a protein database. The result should indicate if your seq and its SNPs fall into protein-coding region(s) and if so, if the SNP alters that protein sequence (non-synonymous) or not (synonymous).