let me tell you the simple logic/science behind it.........
Look my friend it all depend on the size of your template DNA and number of target present in it. genomic DNA from normal human - being diploid - has two copies of each gene.
The size of diploid genome human is around 6 billion bp.
Average size of of nucleotide bp = 660 g/mol
how much DNA from a single cell = 660 x 6 billion
= 3.96 pico gram (pg)
Two target gene copies in 3.96 pg DNA
Though thoerartically, PCR can pick a single copy of target....but thats not the case in reality.
Genrally 25 -100 ng human genomic DNA is recomended for PCR.
how many copies of target gene in 25 ng genomic DNA = (25000pg/3.96pg) x 2
=12626 copies
So around 10,000 - 12000 copies of target DNA are recomended in 25 ul pCR reaction........must be an eye opener for many of us.
So my friend the small is genomic DNA or plasmid the less DNA is required in PCR and vice versa.
Remember number of copies of target DNA is important.
Use above calculation whenever you face similar problem
This is what New England Biolabs recommend at their website:
For low complexity templates (i.e. plasmid, lambda, BAC DNA), use 1 pg–10 ng of DNA per 50 µl reaction
For higher complexity templates (i.e. genomic DNA), use 1 ng–1 µg of DNA per 50 µl reaction
You can see the range is quite broad. Usually for small genomes you would use lower concentrations, since you expect more copies of your target sequence per ng.
To amplify genomic DNA for SNP detection or cloning of fragments we usually use 5 ng in a 10 - 20 µl reaction; for plasmid DNA pg are enough, I usually drop the tip into the plasmid solution without filling it, transfer the tip into the PCR tube & shake it, its always sufficient. You should always try to transfer as little volumen as possible to keep possible inhibitors from the purification procedere on a low level.
Good discussion! I think I should make this question clearer. To get good PCR yield and reduce non-specific products, how much DNA template is generally used and how many cycles do you usually run for a PCR?
You can do a test run with various amounts of DNA 1, 10, 25, 100ng etc (I use 10ng genomic DNA in 25uL PCR, and highly diluted plasmid DNA: about 0.1-1ng) and run various cycle numbers: 30, 35, 40, 45 to find out what works the best for your experiment and primer combination. I also found out that it is important to optimize the annealing temperature for each primer set. You might want to try a temperature gradient (50-70°C) or pick 4 temperatures: at recommended TM and a few degrees above (like 50, 52, 55, 58 °C or so if 50°C is the recommended TM).
I have performed more than 100 reactions with different concentrations. From 10-15ng/25μl reaction I get the optimal results. I have tried it for SNP genotyping,gene expression, STRs, cloning and any other PCR reaction.
I have done PCRs with 5ng genomic DNA. Lower the DNA conc, the easier it is to standardize the reaction. It also saves DNA if you are working on rare specimens.
For genomic DNA we use 50-100ng per 20-50ul reaction. To reduce non specific products is a tad more complex. It is dependent on primers sequences/length as well as enzyme and of course annealing temp used. You could try running touchdown PCR to minimise non specific bands.
We use the Q5 polymerase from NEB and they have their own conditions regarding annealing temp. Its on their website, just input the primer sequences and it gives you the suggested annealing temp for the Q5 reaction. Works very well for genomic PCR
I thought Deng's question had a deeper meaning. The real question, perhaps, is how many "molecules" of the template do you need to obtain a satisfactory and consistent amplification, and whether this value can be generalized across various sequences and mixtures thereof. Plasmids are easy, for you know how many template copies you have on a single plasmid, and how many plasmid molecules you have in a ng. So, One can take a methodical approach in enumerating the basal number of template sequences required for an effective amplication off a plasmid. However, the scenario will be different for the genomic DNA, for you may not know the actual %-representation of your template sequence of interest within the genomic DNA (even as a few minutes of bioinformatics, if with known genomes, can tell you the approaximate value). I have not done such a study, and shall appreciate to know if someone has done it.
Note that for equimolar template concentration, you would need a million times less for plasmid DNA than for gDNA!
That's why, logically, you see people's suggestions are much lower for plasmid DNA, but still very high, just because it's not practical to dilute plasmid DNA that much, and not needed either, since too much template DNA is hardly ever an issue. It can be if a. you need to purify the product downstream and you don't want template DNA to interfere in things like sequencing or cloning. or b. the DNA has impurities like salts that you don't want in your PCR by adding too much template.
On cycle amount. Minor comment. If you want your product fee of errors/mutations, it's better to run fewer cycles with more template. You shouldn't need more than 30 cycles. If it doesn't matter, then by all means, run 35-40cycles.
10-100ng genomic DNA (the high end ONLY if your DNA prep is nicely clean).
1pg-10ng plasmid DNA (the high end ONLY if template plasmid does not interfere downstream)
PS I should hope that you don't approach that 10%vol. of enzyme!
You only need 0.2-0.5Unit (=0.1-0.2ul probably) for a 20ul pcr, that's normally only 1%vol.!
The max 10% vol is is a realistic advice for DNA modifying enzymes such as restriction enzymes, klenow, alkaline phosphatase, ligase, which you use in that range and are sensitive to glycerol.
10 to 100 ng will work perfectly and you will get clear fragments in the gel> But the quality of the DNA will matter a lot. Further the role of other components like MgCl2, Taq, buffer and primers and their optimization is necessary for obtainig the good results
let me tell you the simple logic/science behind it.........
Look my friend it all depend on the size of your template DNA and number of target present in it. genomic DNA from normal human - being diploid - has two copies of each gene.
The size of diploid genome human is around 6 billion bp.
Average size of of nucleotide bp = 660 g/mol
how much DNA from a single cell = 660 x 6 billion
= 3.96 pico gram (pg)
Two target gene copies in 3.96 pg DNA
Though thoerartically, PCR can pick a single copy of target....but thats not the case in reality.
Genrally 25 -100 ng human genomic DNA is recomended for PCR.
how many copies of target gene in 25 ng genomic DNA = (25000pg/3.96pg) x 2
=12626 copies
So around 10,000 - 12000 copies of target DNA are recomended in 25 ul pCR reaction........must be an eye opener for many of us.
So my friend the small is genomic DNA or plasmid the less DNA is required in PCR and vice versa.
Remember number of copies of target DNA is important.
Use above calculation whenever you face similar problem
Usually 1pg–1ng of plasmid DNA template and 1ng–1µg of genomic DNA template
are sufficient for amplification of target DNA using 25-30 PCR cycles.
Also keep in mind that use of high DNA concentration will decrease the specificity of amplicon by resulting in non-specific bands with larger number for cycles. The use of higher DNA concentration can be counter balanced by reducing the number of PCR cycles to increase to specificity.
Generally, for Marker assisted selection and screening of Blast resistant genes in rice, we use 50ng of DNA template for 20μl of reaction mixture and 35 cycle in PCR to get good yield.
we have used as low as 5ng to 50ng /10ul reaction for amplification of genomic DNA. It also depends on the quality of the DNA. U can try doing a gradient(1ng,5ng,10ng,50ng)
I usually start with around 10 ng of a typical plasmid template (assuming 3-5kb), or an equivalent molar quantity of template if not.
One needs to strike a balance between the amount of template, and the degree of amplification that is required to generate an adequate amount of product. Too little template will require more cycles of amplification, and therefore increase the possibility of introducing errors. On the other hand, too much template could result a "dirty" PCR with low yields, and a lot of non-specific amplification. Of course, always make sure that the clone is sequenced.
Around 1-50 ng of Genomic DNA can be used as template in 50 microliter system, provided quality of DNA is good . Plasmdi DNA can be used in picogram, as it contains the cloned gene.
This is good discussion about normal PCR amplification. Can anyone guide about how much soil DNA template (in ng) required for qPCR amplification in case functional gene (mcrA) marker analysis to get good results.
...Remembering that if the PCR product is going to be used directly for cloning then you want to avoid too much contaminating plasmid vector. Of course there is always the option of removing excess vector with a DpnI digest.
Is using 100ng of plasmid DNA too much for pcr reaction using phusion or Q5? I see my backbone in the gel but also able to see my amplified piece. Does anybody know what is the reason behind using 10ng of plasmid DNA? what can go wrong with more DNA? When I dilute to get 10ng my pcr often looks very smeary.
Junaid: I have a disagreement with your calculation:
Since the MW of human diploid DNA in one cell (using 660 g/mole as the estimate of the MW of 1 bp of DNA) is 3.96 x 1012 amu's (not pg), and since 1 amu = 1.66053892 x 10-24 grams, then, 1 human diploid cell would contain ~6.58 picograms dsDNA.
When using "650 g/mole" for the MW of a bp, one calculates ~6.48 pg dsDNA/human diploid cell.
Another value for pg of DNA per human diploid cell has been frequently stated as "6.16 pg DNA/human diploid cell" -- but this would mean that the MW of the bp used in the equation would have to have been 618 g/mole as in the calculation shown at:
So - using your values, but correcting the error in the math, 25 ng of human diploid gDNA would then have ~7604 copies of a dsDNA target. Or, as PCR would actually "see" and act on it: ~15208 single-stranded DNA targets (per 25 ng of dsDNA [single-copy gene] extracted from human diploid cells).
This translates into meaning that 25-100 ng of human diploid ds gDNA per PCReaction would contain 7604-30416 dsDNA target copies of a single-copy gene. Or: in single-stranded target reality, 25-100 ng of human diploid ds gDNA per PCReaction contain 15208-60832 single-stranded target copies, which should translate (at 100% efficiency and a 1-single-strand target copy y-intercept of 38) to Cq values (in qPCR) of 24.1 and 22.1, respectively. For conventional PCR, this would mean that perhaps 25 to 28 cycles would give a nice band on a gel for a single-copy gene in human diploid dsDNA loaded at 25-100 ng/rxn.
1) Generally 50 to 200 ng of DNA I have worked with depending upon the gene of interest, for normal PCR of 16S rDNA gene amplification I have used as little as 50 ng of DNA but for gene specific amplification I have Used as High as 200 ng of DNA.
2) To reduced non specific amplification it all depends upon your primer properties such as its length, its annealing temp., its purity, its conc.
the primer with higher annealing temp generally in the range of 55 to 60 0C are suppose to give less non specific band than that of lower annealing temp.
3) Number of cycle again depend upon the type of Taq pol you are using and the condition of your PCR. Suppose for denaturation temp if you are keeping for 2 to 4 min than in that case max to max you should go for 30 to 35 cycle because beyond that Taq pol activity decreases. Generally you should get good result in 30 to 35 cycle if you are not getting result in 35 cycle then you will not get even if you got for 40 or 45 cycle.
Theory; the volume of PCR reaction should have impact on this. Most of the above answer has omitted that information. Can we assume that generally a PCR reaction would be 10 - 20 ul?
Emperically; I do PCR in 20 ul reaction with 20 - 40 ng of human gDNA to amplify single regions for the purpose of Sanger Sequencing. Running 25 cycles with AmpliSeq 360 mastermix gives high amount of specific product.
Junaid Iqbal gave an interesting answer. The calculation of the weight and the amount of molecules is wrong though. 660 g is a weight of a mole of bp meaning that 6.022 *10^23 bp (Avogadro number) weigh 660 g. Consequently, 6 billion bp in the genome from one cell would weigh (660*6*10^9/6.022*10^23) or 6.58 pg.
660 g per mole is correct for a free nucleotide pair but with the formation of covalent bonds during polymerisation, 2 molecules of water are lost. So, 624 g per mole is better.
And in a diploid human cell (1 genome copy from Mom + one genome copy from Dad), using your values here, the bp are ~6.6 x 109 bp per diploid human genome. So ~12 feet long DNA strand per diploid cell~
Note that "660 g/mole," "650 g/mole" and "618 g/mole" have all been used in these equations... (and with much gusto and good humor as well since 1952 and beyond).
As well: "3.3 x 109 bp"/haploid human genome is a variable figure... depending on whether man or woman (X or Y chromosome content), CNV, SNP frequency, and STR content/individual, satellite sequences, transposon content, viral inception, ALU sequences, GC islands and so on...
So, whatever is accepted as a 'grand average' in any of these values, often amounts to all of our calculations in these regards always being off by about 8%. Which may be significant in some important situations.
In my opinion, the volume of DNA template used is dependent on the type of primers and master mix preparation protocol you are using to set up your PCR. For Viruses, in a 20ul PCR reaction I think using 2ul of the DNA template added to 18 ul of the MM and use a three-step amplification for 45 cycles will yield a better result.
Non-specific bands are due to the quality of primers rather than the cycles. I always used 47 cycles for both conditions and whenever I saw unspecific bands I just design new primers.
I have used from 100 ng -1.5 micrograms in 10 ul as final rxn. It works well. You can use from 30-35 cycles, teoretically if you use more than 40 cycles you can amplify innespecific products or if you need the product to clonning is ok, otherwise with 30 maximum 35. Take into account for you protocol: polymerase amplify 500 bp in 30 seconds or less
10 ng of DNA is ok, potentially, even a lower aliquot or just one molecule of DNA is sufficient. This depends on your protocol/your amplification targets.
I mean, whatever aliquot you decide to use, it is important you standardize your experiments/analysis to a specific starting aliquot.
For example, if you start from cDNA samples sometimes it is difficult to trust on the concentration you measure by nanodrop or by similar instruments, especially to assess non quantitative retro-transcriptase PCR analysis.
In this case, I found useful to run a PCR using primers amplifying an housekeeping gene to see if in some cases your samples contains too much or too less DNA.
By this you can consider to dilute samples providing too strong bands or to re-extract/re-amplify samples providing weak bands before their use in experiments.
I hope this provides further help to previous answers.
I see some imprecisions in the above calculation : avg mass of a nt pair in DNA is 624 (660 - 2 water molecules loss during polymerisation) ;
6 x 10e+9 nt pairs in human genome x 624 g/mole nt pairs = 3744 x 10e+9 g/mole of human genome which divided by Avogadro number gives 624 x 10e-14 g per human genome (6.24 pg). Other part of the calculation seems correct
One can almost do a meta-analysis based on these answers...good:)
A follow-up question: it is rarely possible to keep the similar number of copies of the target DNA in the reaction. How problematic is the large variation of the target DNA among the samples within the plate (provided that this variation is still within the linear range of the assay)?
without any complicated calculations, first, you should measure the concentration and purity of your DNA samples by UV spec. or nano- drop , then, from 20 ng to 100 ng is more sufficient to get good PCR product in 25 microlitre reaction volume...BUT, note that there are many factors affecting on the nonspecific and smear result such as primers, contamination, PCR program and so on..So, you should read PCR troubleshooting to gathering all these notes...on the other hand, PCR product size ,specificity and sharpness is affected by the number of the PCR cycles and controlled by the size of your target ,for example, you should increase your cycles no , annealing , extension times if you have small target and vice versa.
It depends on Quality of DNA and primers and PCR as well;. either you are doing rt PCR or nested. Because different pathogens have different annealing temperature and different standardization parameters.
If DNA quality is good than 10-100ng per microlitre is okay. For good pcr band it is necessary to maintain appropriate MgCl2 concentration. To avoid smearing and other bad quality of pcr product need standardization of pcr reaction like error and trial basis. Thanks.
I routinely use 100 ng of gDNA and 1-10ng of plasmid DNA in PCR reactions. However, something amount of added DNA also depends on specific need. I agree with one of the previous comment that less is good. Suppose you infected some cell-line with CRISPR library and want to amplify targets subsequently. In this case, I generally use 2 ug of gDNA per reaction and in triplicates to cover all the library complexity. I never faced any issue using 2 ug of gDNA per reaction. If you want to run a highly sensitive digital PCR assay then 1ng of plasmid DNA may also be more than enough. You may need to dilute it 10^2-10^4 in that case.
1-10ng of gDNA template and 1-10ng of plasmid DNA in PCR reactions. in a 50uL reaction is fairly standard. More is not good. Although amplification can lower starting amounts is feasible, it is not recommended for cloning as you increase the likelihood of PCR errors with more extensive amplification.
10-100 ng gDNA and 1-10 ng of plasmid DNA in 50 uL Rxn is required for getting good results. I always use in that range only for gDNA and plasmid DNA respectively.
The concentration of DNA template depends on the source. Normally used concentration are 100-250 ng for mammalian genomic DNA and 20 ng for linearized plasmid DNA (circular plasmid DNA is slightly less efficiently amplified) per 50µl reaction.