As others have mentioned, highly multiplexed PCR is notoriously difficult. Two main reasons are: possible primer-dimers among any combination of the 44 primers you mix and unspecific binding and product generation again among any combination of the 44 primers you have.

Quite often the only reason a primer pair is specific is because only the target sequence gives a resonable product size while still both primers will bind unspecifically at other regions of the genome, maybe because there are only mismatches in the 5'-half of the primer. But then they don't have a complementary primer binding site for the second primer close by. The more primers you mix, the more likely it becomes that primer number 43 will eventually unspecifically bind 600 bp away antisense to an unspecific binding of primer number 1 sense. Get the picture? You can use ncbi Primer-Blast and see how specific two primers really are....Doing this manually with combinatorial 44 primers is a long task......

I have not tried it, but MPprimer seems to be a nice tool to design multiplex primers. It checks both for primer dimers and unspecific genomewide binding using MFEprimer.

Highly-multiplexed PCR is possible, as the PCR based enrichment algorithms have shown in gene-panel NGS. But it is not trivial and doesn't always work 100% on first attempt.

http://biocompute.bmi.ac.cn/MPprimer/

Maran,

Seconding what Mark has already said, 44 primers is way too much to add to a PCR tube at once. It is not at all surprising that you are just getting a smear when you run the product since so many of these primers being in the tube will lead to unspecific amplification. 

Depending on your product sizes, you could split these primers into sets of 10 or 15 primers each so that you can get the desired amplification and also be able to analyse results with ease. I have seen people put 18 primers in a tube and the results are not very clearly distinguishable. So, my advice would be to stick to sets of 10-12 primers. This way, even in case you to repeat a PCR for clarification, you just need to repeat a particular set and not the whole reaction. 

Thanks for all the feedback!

I have another reaction in which I use 40 primers and it works perfectly, this is the reason I think 44 primers should theoretically work. My product size is about 500bp and my primers range from 25-40bp. Not only are there a lot of them they are large primers. The reason they are so large is because of the attached adapter sequence that gets incorporated into my target sequence after the first round of PCR. If the primers are priming off each other this would account for the smear I see below my target sequence but I don't understand why there is a smear above my target sequence. Also, I've validated all of the primers individually and sequences the resulting PCR products and they all prime their intended sequence, so why when they're all together would they suddenly become non-specific? 

Maybe primers align to each other? Do you get no-product in no-template reaction?

As others have mentioned, highly multiplexed PCR is notoriously difficult. Two main reasons are: possible primer-dimers among any combination of the 44 primers you mix and unspecific binding and product generation again among any combination of the 44 primers you have.

Quite often the only reason a primer pair is specific is because only the target sequence gives a resonable product size while still both primers will bind unspecifically at other regions of the genome, maybe because there are only mismatches in the 5'-half of the primer. But then they don't have a complementary primer binding site for the second primer close by. The more primers you mix, the more likely it becomes that primer number 43 will eventually unspecifically bind 600 bp away antisense to an unspecific binding of primer number 1 sense. Get the picture? You can use ncbi Primer-Blast and see how specific two primers really are....Doing this manually with combinatorial 44 primers is a long task......

I have not tried it, but MPprimer seems to be a nice tool to design multiplex primers. It checks both for primer dimers and unspecific genomewide binding using MFEprimer.

Highly-multiplexed PCR is possible, as the PCR based enrichment algorithms have shown in gene-panel NGS. But it is not trivial and doesn't always work 100% on first attempt.

http://biocompute.bmi.ac.cn/MPprimer/

To answer your question regarding the non-specific amplification above the expected/desired fragments, it is quite reasonable since by introducing 44 pairs of primers you have to check for all the possible combinations of these primers and see if you are going to get any undesired amplification (against the appropriate genomic or transcript DNA database, depending on your template). I expect you will. I would try the software suggested by Lars-Erik Wehner, (although I haven't used it myself either) because going through all of the primer combinations manually seems like a nightmare

Primer-blast is a good tool when you search the specificity for your primer pairs...

I can suggest you to try using, for example, 20-30 primers and to do a progressive scale increasing the number of primers used. You said that you tried to use at the same time 40 primers...they were in common with the current 44 primers?? If the answer is yes, firstly try to add only 1 primer at time.

The smear that you see below your ipotetically band could be caused by a degradation of your DNA or a degradation of your primer... the annealing primer is the same/ nearly similar for all primers or not? Perhaps by changing the annealing temperature you could obtain the band awaited!

Good luck!

By overcoming primer Dimer formation,u may have solution to ur problem.

I guess you might trying to do 5C (Chromosome Conformation Capture Carbon Copy), since this method requires to do PCR with many primers at the same time. As others have mentioned, specific primer combination might case smear. So you need to find out which combination has problem by computationally or experimentally.

For the experimentally, how about try several groups of  20-30 primers set and find out which combination cause smear.