Ten quick tips for building FAIR workflows.

Research data is accumulating rapidly and with it the challenge of fully reproducible science. As a consequence, implementation of high-quality management of scientific data has become a global priority. The FAIR (Findable, Accesible, Interoperable and Reusable) principles provide practical guidelines for maximizing the value of research data; however, processing data using workflows-systematic executions of a series of computational tools-is equally important for good data management. The FAIR principles have recently been adapted to Research Software (FAIR4RS Principles) to promote the reproducibility and reusability of any type of research software. Here, we propose a set of 10 quick tips, drafted by experienced workflow developers that will help researchers to apply FAIR4RS principles to workflows. The tips have been arranged according to the FAIR acronym, clarifying the purpose of each tip with respect to the FAIR4RS principles. Altogether, these tips can be seen as practical guidelines for workflow developers who aim to contribute to more reproducible and sustainable computational science, aiming to positively impact the open science and FAIR community.

Detection of Fusion Genes to Determine Minimal Residual Disease in Leukemia Using Next-Generation Sequencing.

BACKGROUND: Measuring minimal residual disease (MRD), the persistence of leukemic cells after treatment, is important for monitoring leukemia recurrence. The current methods for monitoring MRD are flow cytometry, to assess aberrant immune phenotypes, and digital droplet PCR (ddPCR), to target genetic aberrations such as single-nucleotide variants and gene fusions. We present the performance of an RNA-based next-generation sequencing (NGS) method for MRD gene fusion detection compared with ddPCR. This method may have advantages, including the capacity to analyze different genetic aberrations and patients in 1 experiment. In particular, detection at the RNA level may be highly sensitive if the genetic aberration is highly expressed. METHODS: We designed a probe-based NGS panel targeting the breakpoints of 11 fusion genes previously identified in clinical patients and 2 fusion genes present in cell lines. Blocking probes were added to prevent nonspecific enrichment. Each patient RNA sample was diluted in background RNA, depleted for rRNA and globin mRNA, converted to cDNA, and prepared for sequencing. Unique sequence reads, identified by unique molecular identifiers, were aligned directly to reference transcripts. The same patient and cell-line samples were also analyzed with ddPCR for direct comparison. RESULTS: Our NGS method reached a maximum sensitivity of 1 aberrant cell in 10 000 cells and was mostly within a factor of 10 compared with ddPCR. CONCLUSIONS: Our detection limit was below the threshold of 1:1000 recommended by European Leukemia Net. Further optimizations are easy to implement and are expected to boost the sensitivity of our method to diagnostically obtained ddPCR thresholds.

Targeted RNA-Sequencing Enables Detection of Relevant Translocations and Single Nucleotide Variants and Provides a Method for Classification of Hematological Malignancies-RANKING.

BACKGROUND: Patients with hematological malignancies (HMs) carry a wide range of chromosomal and molecular abnormalities that impact their prognosis and treatment. Since no current technique can detect all relevant abnormalities, technique(s) are chosen depending on the reason for referral, and abnormalities can be missed. We tested targeted transcriptome sequencing as a single platform to detect all relevant abnormalities and compared it to current techniques. MATERIAL AND METHODS: We performed RNA-sequencing of 1385 genes (TruSight RNA Pan-Cancer, Illumina) in bone marrow from 136 patients with a primary diagnosis of HM. We then applied machine learning to expression profile data to perform leukemia classification, a method we named RANKING. Gene fusions for all the genes in the panel were detected, and overexpression of the genes EVI1, CCND1, and BCL2 was quantified. Single nucleotide variants/indels were analyzed in acute myeloid leukemia (AML), myelodysplastic syndrome and patients with acute lymphoblastic leukemia (ALL) using a virtual myeloid (54 genes) or lymphoid panel (72 genes). RESULTS: RANKING correctly predicted the leukemia classification of all AML and ALL samples and improved classification in 3 patients. Compared to current methods, only one variant was missed, c.2447A>T in KIT (RT-PCR at 10-4), and BCL2 overexpression was not seen due to a t(14; 18)(q32; q21) in 2% of the cells. Our RNA-sequencing method also identified 6 additional fusion genes and overexpression of CCND1 due to a t(11; 14)(q13; q32) in 2 samples. CONCLUSIONS: Our combination of targeted RNA-sequencing and data analysis workflow can improve the detection of relevant variants, and expression patterns can assist in establishing HM classification.

A next-generation sequencing method for gene doping detection that distinguishes low levels of plasmid DNA against a background of genomic DNA.

Gene doping confers health risks for athletes and is a threat to fair competition in sports. Therefore the anti-doping community has given attention on its detection. Previously published polymerase chain reaction-based methodologies for gene doping detection are targeting exon-exon junctions in the intron-less transgene. However, because these junctions are known, it would be relatively easy to evade detection by tampering with the copyDNA sequences. We have developed a targeted next-generation sequencing based assay for the detection of all exon-exon junctions of the potential doping genes, EPO, IGF1, IGF2, GH1, and GH2, which is resistant to tampering. Using this assay, all exon-exon junctions of copyDNA of doping genes could be detected with a sensitivity of 1296 copyDNA copies in 1000 ng of genomic DNA. In addition, promotor regions and plasmid-derived sequences are readily detectable in our sequence data. While we show the reliability of our method for a selection of genes, expanding the panel to detect other genes would be straightforward. As we were able to detect plasmid-derived sequences, we expect that genes with manipulated junctions, promotor regions, and plasmid or virus-derived sequences will also be readily detected.

NIPTeR: an R package for fast and accurate trisomy prediction in non-invasive prenatal testing.

BACKGROUND: Various algorithms have been developed to predict fetal trisomies using cell-free DNA in non-invasive prenatal testing (NIPT). As basis for prediction, a control group of non-trisomy samples is needed. Prediction accuracy is dependent on the characteristics of this group and can be improved by reducing variability between samples and by ensuring the control group is representative for the sample analyzed. RESULTS: NIPTeR is an open-source R Package that enables fast NIPT analysis and simple but flexible workflow creation, including variation reduction, trisomy prediction algorithms and quality control. This broad range of functions allows users to account for variability in NIPT data, calculate control group statistics and predict the presence of trisomies. CONCLUSION: NIPTeR supports laboratories processing next-generation sequencing data for NIPT in assessing data quality and determining whether a fetal trisomy is present. NIPTeR is available under the GNU LGPL v3 license and can be freely downloaded from https://github.com/molgenis/NIPTeR or CRAN.

Genetic Screening Test to Detect Translocations in Acute Leukemias by Use of Targeted Locus Amplification.

BACKGROUND: Over 500 translocations have been identified in acute leukemia. To detect them, most diagnostic laboratories use karyotyping, fluorescent in situ hybridization, and reverse transcription PCR. Targeted locus amplification (TLA), a technique using next-generation sequencing, now allows detection of the translocation partner of a specific gene, regardless of its chromosomal origin. We present a TLA multiplex assay as a potential first-tier screening test for detecting translocations in leukemia diagnostics. METHODS: The panel includes 17 genes involved in many translocations present in acute leukemias. Procedures were optimized by using a training set of cell line dilutions and 17 leukemia patient bone marrow samples and validated by using a test set of cell line dilutions and a further 19 patient bone marrow samples. Per gene, we determined if its region was involved in a translocation and, if so, the translocation partner. To balance sensitivity and specificity, we introduced a gray zone showing indeterminate translocation calls needing confirmation. We benchmarked our method against results from the 3 standard diagnostic tests. RESULTS: In patient samples passing QC, we achieved a concordance with benchmarking tests of 81% in the training set and 100% in the test set, after confirmation of 4 and nullification of 3 gray zone calls (in total). In cell line dilutions, we detected translocations in 10% aberrant cells at several genetic loci. CONCLUSIONS: Multiplex TLA shows promising results as an acute leukemia screening test. It can detect cryptic and other translocations in selected genes. Further optimization may make this assay suitable for diagnostic use.

CoNVaDING: Single Exon Variation Detection in Targeted NGS Data.

We have developed a tool for detecting single exon copy-number variations (CNVs) in targeted next-generation sequencing data: CoNVaDING (Copy Number Variation Detection In Next-generation sequencing Gene panels). CoNVaDING includes a stringent quality control (QC) metric, that excludes or flags low-quality exons. Since this QC shows exactly which exons can be reliably analyzed and which exons are in need of an alternative analysis method, CoNVaDING is not only useful for CNV detection in a research setting, but also in clinical diagnostics. During the validation phase, CoNVaDING detected all known CNVs in high-quality targets in 320 samples analyzed, giving 100% sensitivity and 99.998% specificity for 308,574 exons. CoNVaDING outperforms existing tools by exhibiting a higher sensitivity and specificity and by precisely identifying low-quality samples and regions.

Excretion of radionuclides in human breast milk after nuclear medicine examinations. Biokinetic and dosimetric data and recommendations on breastfeeding interruption.

PURPOSE: To review early recommendations and propose guidelines for breastfeeding interruption after administration of radiopharmaceuticals, based on additional biokinetic and dosimetric data. METHODS: Activity concentrations in breast milk from 53 breastfeeding patients were determined. The milk was collected at various times after administration of 16 different radiopharmaceuticals. The fraction of the activity administered to the mother excreted in the breast milk, the absorbed doses to various organs and tissues and the effective dose to the infant were estimated. RESULTS: The fraction of the administered activity excreted per millilitre of milk varied widely from 10(-10) to 10(-3) MBq/MBq administered. For (99m)Tc-labelled radiopharmaceuticals, the total fraction of the administered activity excreted in the milk varied from 0.0057 % for (99m)Tc-labelled red blood cells (RBC) to 19 % for (99m)Tc-pertechnetate. The effective dose to an infant per unit activity administered to the mother ranged from 6.7 × 10(-6) mSv/MBq for (99m)Tc-labelled RBC to 3.6 × 10(-2) mSv/MBq for (99m)Tc-pertechnetate. For the other radiopharmaceuticals, the total fraction of administered activity excreted in the milk varied from 0.018 % ((51)Cr-EDTA) to 48 % ((131)I-NaI). The effective dose ranged from 5.6 × 10(-5) mSvinfant/MBqmother ((51)Cr-EDTA) to 106 mSvinfant/MBqmother ((131)I-NaI). CONCLUSIONS: Based on an effective dose limit of 1 mSv to the infant and a typical administered activity, we recommend cessation of breastfeeding for (131)I-NaI and interruption of feeding for 12 h for (125)I-iodohippurate, (131)I-iodohippurate, (99m)Tc-pertechnetate and (99m)Tc-MAA. During this 12-h period all breast milk should be expressed at least three times and discarded. For the other radiopharmaceuticals included in this study, no interruption of breastfeeding is necessary.

Macromolecular crowding effects on two homologs of ribosomal protein s16: protein-dependent structural changes and local interactions.

Proteins function in cellular environments that are crowded with biomolecules, and in this reduced available space, their biophysical properties may differ from those observed in dilute solutions in vitro. Here, we investigated the effects of a synthetic macromolecular crowding agent, dextran 20, on the folded states of hyperthermophilic (S16Thermo) and mesophilic (S16Meso) homologs of the ribosomal protein S16. As expected for an excluded-volume effect, the resistance of the mesophilic protein to heat-induced unfolding increased in the presence of dextran 20, and chemical denaturation experiments at different fixed temperatures showed the macromolecular crowding effect to be temperature-independent. Förster resonance energy transfer experiments show that intramolecular distances between an intrinsic Trp residue and BODIPY-labeled S16Meso depend on the level of the crowding agent. The BODIPY group was attached at three specific positions in S16Meso, allowing measurements of three intraprotein distances. All S16Meso variants exhibited a decrease in the average Trp-BODIPY distance at up to 100 mg/mL dextran 20, whereas the changes in distance became anisotropic (one distance increased, two distances decreased) at higher dextran concentrations. In contrast, the two S16Thermo mutants did not show any changes in Trp-BODIPY distances upon increase of dextran 20 concentrations. It should be noted that the fluorescence quantum yields and lifetimes of BODIPY attached to the two S16 homologs decreased gradually in the presence of dextran 20. To investigate the origin of this decrease, we studied the BODIPY quantum yield in three protein variants in the presence of a tyrosine-labeled dextran. The experiments revealed distinct tyrosine quenching behaviors of BODIPY in the three variants, suggesting a dynamic local interaction between dextran and one particular S16 variant.

Comparing Sequence-Based and Literature-Based Pathogenicity Scoring Methods for Human Variants.

Assessing the pathogenicity of genetic variants is a critical aspect of genomic medicine and precision healthcare. Over the last decades, the identification of genetic variants and their characterization has become simpler (advent of high-throughput sequencing technologies, analysis, and visualization support tools, etc.). However, the quality of assessments to distinguish benign from pathogenic variants is critical to inform clinical decision-making and improve patient outcomes. In this article, we investigate the relationships using correlation tests between the characterization of genetic variants in the literature and their pathogenicity scores computed by two state-of-the-art assessment tools (SIFT and PolyPhen-2).

Cas9-directed long-read sequencing to resolve optical genome mapping findings in leukemia diagnostics.

Leukemias are genetically heterogeneous and diagnostics therefore includes various standard-of-care (SOC) techniques, including karyotyping, SNP-array and FISH. Optical genome mapping (OGM) may replace these as it detects different types of structural aberrations simultaneously and additionally detects much smaller aberrations (500 bp vs 5-10 Mb with karyotyping). However, its resolution may still be too low to define clinical relevance of aberrations when they are located between two OGM labels or when labels are not distinct enough. Here, we test the potential of Cas9-directed long-read sequencing (LRS) as an additional technique to resolve such potentially relevant new findings. From an internal Bionano implementation study we selected ten OGM calls that could not be validated with SOC methods. Per variant we designed crRNAs for Cas9 enrichment, prepared libraries and sequenced them on a MinION/GridION device. We could confirm all aberrations and, importantly, the actual breakpoints of the OGM calls were located between 0.2 and 5.5 kb of the OGM-estimated breakpoints, confirming the high reliability of OGM. Furthermore, we show examples of redefinition of aberrations between labels that enable judgment of clinical relevance. Our results suggest that Cas9-directed LRS can be a relevant and flexible secondary technique in diagnostic workflows including OGM.

Future of Dutch NGS-Based Newborn Screening: Exploring the Technical Possibilities and Assessment of a Variant Classification Strategy.

In this study, we compare next-generation sequencing (NGS) approaches (targeted panel (tNGS), whole exome sequencing (WES), and whole genome sequencing (WGS)) for application in newborn screening (NBS). DNA was extracted from dried blood spots (DBS) from 50 patients with genetically confirmed inherited metabolic disorders (IMDs) and 50 control samples. One hundred IMD-related genes were analyzed. Two data-filtering strategies were applied: one to detect only (likely) pathogenic ((L)P) variants, and one to detect (L)P variants in combination with variants of unknown significance (VUS). The variants were filtered and interpreted, defining true/false positives (TP/FP) and true/false negatives (TN/FN). The variant filtering strategies were assessed in a background cohort (BC) of 4833 individuals. Reliable results were obtained within 5 days. TP results (47 patient samples) for tNGS, WES, and WGS results were 33, 31, and 30, respectively, using the (L)P filtering, and 40, 40, and 38, respectively, when including VUS. FN results were 11, 13, and 14, respectively, excluding VUS, and 4, 4, and 6, when including VUS. The remaining FN were mainly samples with a homozygous VUS. All controls were TN. Three BC individuals showed a homozygous (L)P variant, all related to a variable, mild phenotype. The use of NGS-based workflows in NBS seems promising, although more knowledge of data handling, automated variant interpretation, and costs is needed before implementation.

An interconnected data infrastructure to support large-scale rare disease research.

The Solve-RD project brings together clinicians, scientists, and patient representatives from 51 institutes spanning 15 countries to collaborate on genetically diagnosing ("solving") rare diseases (RDs). The project aims to significantly increase the diagnostic success rate by co-analyzing data from thousands of RD cases, including phenotypes, pedigrees, exome/genome sequencing, and multiomics data. Here we report on the data infrastructure devised and created to support this co-analysis. This infrastructure enables users to store, find, connect, and analyze data and metadata in a collaborative manner. Pseudonymized phenotypic and raw experimental data are submitted to the RD-Connect Genome-Phenome Analysis Platform and processed through standardized pipelines. Resulting files and novel produced omics data are sent to the European Genome-Phenome Archive, which adds unique file identifiers and provides long-term storage and controlled access services. MOLGENIS "RD3" and Café Variome "Discovery Nexus" connect data and metadata and offer discovery services, and secure cloud-based "Sandboxes" support multiparty data analysis. This successfully deployed and useful infrastructure design provides a blueprint for other projects that need to analyze large amounts of heterogeneous data.

Direct observation of protein unfolded state compaction in the presence of macromolecular crowding.

Proteins fold and function in cellular environments that are crowded with other macromolecules. As a consequence of excluded volume effects, compact folded states of proteins should be indirectly stabilized due to destabilization of extended unfolded conformations. Here, we assess the role of excluded volume in terms of protein stability, structural dimensions and folding dynamics using a sugar-based crowding agent, dextran 20, and the small ribosomal protein S16 as a model system. To specifically address dimensions, we labeled the protein with BODIPY at two positions and measured Trp-BODIPY distances under different conditions. As expected, we found that dextran 20 (200 mg/ml) stabilized the variants against urea-induced unfolding. At conditions where the protein is unfolded, Förster resonance energy transfer measurements reveal that in the presence of dextran, the unfolded ensemble is more compact and there is residual structure left as probed by far-ultraviolet circular dichroism. In the presence of a crowding agent, folding rates are faster in the two-state regime, and at low denaturant concentrations, a kinetic intermediate is favored. Our study provides direct evidence for protein unfolded-state compaction in the presence of macromolecular crowding along with its energetic and kinetic consequences.

Targeted next-generation sequencing can replace Sanger sequencing in clinical diagnostics.

Mutation detection through exome sequencing allows simultaneous analysis of all coding sequences of genes. However, it cannot yet replace Sanger sequencing (SS) in diagnostics because of incomplete representation and coverage of exons leading to missing clinically relevant mutations. Targeted next-generation sequencing (NGS), in which a selected fraction of genes is sequenced, may circumvent these shortcomings. We aimed to determine whether the sensitivity and specificity of targeted NGS is equal to those of SS. We constructed a targeted enrichment kit that includes 48 genes associated with hereditary cardiomyopathies. In total, 84 individuals with cardiomyopathies were sequenced using 151 bp paired-end reads on an Illumina MiSeq sequencer. The reproducibility was tested by repeating the entire procedure for five patients. The coverage of ≥30 reads per nucleotide, our major quality criterion, was 99% and in total ∼21,000 variants were identified. Confirmation with SS was performed for 168 variants (155 substitutions, 13 indels). All were confirmed, including a deletion of 18 bp and an insertion of 6 bp. The reproducibility was nearly 100%. We demonstrate that targeted NGS of a disease-specific subset of genes is equal to the quality of SS and it can therefore be reliably implemented as a stand-alone diagnostic test.

Successful noninvasive trisomy 18 detection using single molecule sequencing.

BACKGROUND: Noninvasive trisomy 21 detection performed by use of massively parallel sequencing is achievable with high diagnostic sensitivity and low false-positive rates. Detection of fetal trisomy 18 and 13 has been reported as well but seems to be less accurate with the use of this approach. The reduced accuracy can be explained by PCR-introduced guanine-cytosine (GC) bias influencing sequencing data. Previously, we demonstrated that sequence data generated by single molecule sequencing show virtually no GC bias and result in a more pronounced noninvasive detection of fetal trisomy 21. In this study, single molecule sequencing was used for noninvasive detection of trisomy 18 and 13. METHODS: Single molecule sequencing was performed on the Helicos platform with free DNA isolated from maternal plasma from 11 weeks of gestation onward (n = 17). Relative sequence tag density ratios were calculated against male control plasma samples and results were compared to those of previous karyotyping. RESULTS: All trisomy 18 fetuses were identified correctly with a diagnostic sensitivity and specificity of 100%. However, low diagnostic sensitivity and specificity were observed for fetal trisomy 13 detection. CONCLUSIONS: We successfully applied single molecule sequencing in combination with relative sequence tag density calculations for noninvasive trisomy 18 detection using free DNA from maternal plasma. However, noninvasive trisomy 13 detection was not accurate and seemed to be influenced by more than just GC content.

Synthesis of fluorescent ring-fused 2-pyridone peptidomimetics.

Thiazolino fused 2-pyridone peptidomimetics are of significant biological importance due to their ability to interfere with adhesive fiber formation in uropathogenic Escherichia coli and oligomerization of amyloid fibers. We have developed an efficient synthetic route to fluorescent BODIPY analogues, with structural diversification from a key intermediate enabling introduction of C-2 substituents and late incorporation of the BODIPY moiety. A mild lithium halide mediated hydrolysis enabled preparation of peptidomimetic fluorophores with useful photophysical properties for further chemical biology applications.

Cross-camera comparison of ROI-based semi-quantitative ¹²³I-IBZM SPECT data in healthy volunteers using an anthropomorphic phantom for calibration.

BACKGROUND: In (123)I-Iolopride (IBZM) SPECT reference values may diverge between camera systems. If multicenter pooling of normal material databases is needed, differences in measured semi-quantitative data due to equipment performance and reconstruction parameters have to be investigated in each instance to determine the comparability. PURPOSE: To explore the differences in (123)I-IBZM measured uptake ratios between two different gamma cameras in healthy controls, the intra-rater reproducibility of the image evaluation method and the possibility to equalize uptake ratios by calibration through an anthropomorphic phantom. MATERIAL AND METHODS: Differences in ROI-based semi-quantitative data from two different gamma camera systems, the three-headed brain dedicated Neurocam and the two-headed multipurpose hybrid system Infinia Hawkeye, were studied using image data from a group of healthy volunteers and an anthropomorphic brain-phantom scanned with both cameras. Several reconstruction methods and corrections were applied. To test the reliability of the ROI method, the intra-observer reproducibility was determined for the ROI method in this study. RESULTS: The ROI method had a high reliability. Differences in mean measured uptake (123)I-IBZM ratios in healthy controls varied between 2.9% and 6.5% depending on reconstruction and correction for attenuation and scatter. After calibration, the differences decreased. There were no statistically significant differences between corrected ratios from the two camera systems in the study when images were reconstructed with attenuation correction. CONCLUSION: The conformity of uptake ratios in attenuation corrected (123)I-IBZM images derived from the two different cameras was improved by using an anthropomorphic phantom for calibration.

Curation and expansion of the Human Phenotype Ontology for systemic autoinflammatory diseases improves phenotype-driven disease-matching.

Introduction: Accurate and standardized phenotypic descriptions are essential in diagnosing rare diseases and discovering new diseases, and the Human Phenotype Ontology (HPO) system was developed to provide a rich collection of hierarchical phenotypic descriptions. However, although the HPO terms for inborn errors of immunity have been improved and curated, it has not been investigated whether this curation improves the diagnosis of systemic autoinflammatory disease (SAID) patients. Here, we aimed to study if improved HPO annotation for SAIDs enhanced SAID identification and to demonstrate the potential of phenotype-driven genome diagnostics using curated HPO terms for SAIDs. Methods: We collected HPO terms from 98 genetically confirmed SAID patients across eight different European SAID expertise centers and used the LIRICAL (Likelihood Ratio Interpretation of Clinical Abnormalities) computational algorithm to estimate the effect of HPO curation on the prioritization of the correct SAID for each patient. Results: Our results show that the percentage of correct diagnoses increased from 66% to 86% and that the number of diagnoses with the highest ranking increased from 38 to 45. In a further pilot study, curation also improved HPO-based whole-exome sequencing (WES) analysis, diagnosing 10/12 patients before and 12/12 after curation. In addition, the average number of candidate diseases that needed to be interpreted decreased from 35 to 2. Discussion: This study demonstrates that curation of HPO terms can increase identification of the correct diagnosis, emphasizing the high potential of HPO-based genome diagnostics for SAIDs.

Dynamics and size of cross-linking-induced lipid nanodomains in model membranes.

Changes of membrane organization upon cross-linking of its components trigger cell signaling response to various exogenous factors. Cross-linking of raft gangliosides GM1 with cholera toxin (CTxB) was shown to cause microscopic phase separation in model membranes, and the CTxB-GM1 complexes forming a minimal lipid raft unit are the subject of ongoing cell membrane research. Yet, those subdiffraction sized rafts have never been described in terms of size and dynamics. By means of two-color z-scan fluorescence correlation spectroscopy, we show that the nanosized domains are formed in model membranes at lower sphingomyelin (Sph) content than needed for the large-scale phase separation and that the CTxB-GM1 complexes are confined in the domains poorly stabilized with Sph. Förster resonance energy transfer together with Monte Carlo modeling of the donor decay response reveal the domain radius of ~8 nm, which increases at higher Sph content. We observed two types of domains behaving differently, which suggests a dual role of the cross-linker: first, local transient condensation of the GM1 molecules compensating for a lack of Sph and second, coalescence of existing nanodomains ending in large-scale phase separation.