Can a genome-wide SNP chip be used for a case-control study if the DNA concentration is too low for processing?

For cases of 10-20 ng/μL with a total amount of less than 1 μg, it is not recommended to directly run the standard protocol on the instrument. First, check A260/A280 and electrophoresis to confirm that the DNA integrity has not undergone severe degradation. If the concentration is simply low and impurities are minimal, vacuum centrifugal concentration can be used to bring it to around 50 ng/μL, but the concentration time should not be too long, to avoid salts and EDTA also being concentrated and affecting digestion efficiency. The problem of a total amount of less than 1 μg cannot be solved by concentration; re-extraction or combining extraction products from the same sample is required. If it is a rare sample that cannot be re-extracted, whole genome amplification can be considered, but amplified samples and non-amplified samples must be randomly distributed across different batches, and amplification must be included as a covariate in the statistical analysis. First select 5-8 low-concentration samples for a small-scale instrument run to see whether the call rate is within the platform's acceptable range, then decide on the large-batch plan.

For 200 case-control samples, it is recommended to retain at least 5-6 replicate samples, approximately 2-3%. These can be the same DNA aliquoted into two tubes labeled with different numbers, or 3 cases and 3 controls can be selected for technical replicates. Place 1-2 blank controls per plate, generally with TE or elution buffer added, without performing WGA, to monitor cross-plate contamination. Additionally, place 1-2 standard reference DNAs such as NA12878 or Coriell cell line DNA to monitor chip batch and scanner drift. Do not concentrate low-concentration samples on a single plate, as this can easily produce plate effects. During data analysis, first use PLINK or GenomeStudio for call rate filtering, removing samples with less than 95% and SNP loci with less than 98%, then check whether the consistency of replicate samples reaches 99% or above.

Dilution does not increase the total amount, it only changes the concentration. For example, if you dilute 50 ng/μL to 25 ng/μL, and the total amount was already insufficient, it still will not meet the requirements for loading onto the instrument. If concentration is performed by ethanol precipitation, it may co-precipitate proteins, RNA, or salts, leading to falsely elevated OD measurements, reduced digestion or hybridization efficiency, and during genotyping, allelic dropout or cluster scattering may easily occur. A better approach is to use Qubit or PicoGreen for true dsDNA quantification, rather than relying solely on Nanodrop. If the A260/A280 after concentration deviates significantly from 1.8-2.0, purify first and then concentrate. For 200 samples, it is recommended to record the extraction batch, concentration, and concentration method for each sample when processing in groups, and later during quality control, troubleshoot abnormal samples based on these variables.