Target amplification
In 77 infertile males analysed, 109 rare protein coding DNMs were
identified and PCR amplified targeted phasing performed, for a basic
schematic of the methods see Supplementary Figure 1. Irrespective of
known iSNP presence, primers were designed for the target regions
(determined as per Figure 1.b) of all 109 DNMs (Supplementary Table 1)
using our in-house pipeline which designs primers using Primer3 version
2.3.6 (Untergasser et al., 2012) and Human Genome Reference Assembly
GCRh37. Automated rounds of primer creation were run for each target,
selecting the most optimal primer pair based on predicted fragment size,
self-binding, hairpin factors, reverse primer to forward primer binding,
and non-specific genome binding. All expected fragment sizes were
limited to a maximum length of 12 kb for greater enrichment success and
speed, though optimisation steps
were required for 50 % of the targets (Supplementary Figure 2).
DNMs without exome supported iSNPs within a 10 kb region were selected
for intronic iSNP searching within 5kb (Figure 1.b3). For this group
primers were designed in 5 kb windows around the DNM. iSNPs from the
~5 kb region were identified in the introns of parental
ONT data in the downstream bioinformatics. Long-range PCR target
enrichment was carried out using optimised running conditions of 2
separate supermixes/enzymes, BioRad iproof and TakaraBio PrimeSTAR GLX
(Supplementary Table 2). Primary steps use dilution to reduce
contaminants and inhibitors, where dilution was not enough a hotstart
supermix (Quantabio RepliQa HiFi ToughMix) was used which reduces the
need for annealing temperature specificity (Supplementary Table 2).
The number of primer pairs that
required bespoke optimisation was reduced to just 12% of the samples by
using broad and basic steps as per the Supplementary Figure 2b. Only 5
% of the primer pairs required re-designing, the rest were successfully
optimised across the first 5 stages in supplementary Figure 1b.
Supplementary Figure 1b steps were chosen based on simple and effective
PCR optimisation that could be easily applied on a broad scale for a
large number of samples. These measures gave a long-range PCR success
rate of >80%, reducing the time required for primer
specific optimisation and/or re-design. Sample fragment sizes were
confirmed using gel electrophoresis, and quantities were measured with
the Qubit dsDNA HS kit (Thermo Fisher Scientific, Waltham, MA, USA),
with the best quality supermix enrichment for each given sample/target
selected for sequencing.