Browsing by Subject "FRACTION"

Sort by: Order: Results:

Now showing items 1-8 of 8
  • Teder, Hindrek; Paluoja, Priit; Rekker, Kadri; Salumets, Andres; Krjutškov, Kaarel; Palta, Priit (2019)
    Non-invasive prenatal testing (NIPT) enables accurate detection of fetal chromosomal trisomies. The majority of publicly available computational methods for sequencing-based NIPT analyses rely on low-coverage whole-genome sequencing (WGS) data and are not applicable for targeted high-coverage sequencing data from cell-free DNA samples. Here, we present a novel computational framework for a targeted high-coverage sequencing-based NIPT analysis. The developed framework uses a hidden Markov model (HMM) in conjunction with a supplemental machine learning model, such as decision tree (DT) or support vector machine (SVM), to detect fetal trisomy and parental origin of additional fetal chromosomes. These models were developed using simulated datasets covering a wide range of biologically relevant scenarios with various chromosomal quantities, parental origins of extra chromosomes, fetal DNA fractions, and sequencing read depths. Developed models were tested on simulated and experimental targeted sequencing datasets. Consequently, we determined the functional feasibility and limitations of each proposed approach and demonstrated that read count-based HMM achieved the best overall classification accuracy of 0.89 for detecting fetal euploidies and trisomies on simulated dataset. Furthermore, we show that by using the DT and SVM on the HMM classification results, it was possible to increase the final trisomy classification accuracy to 0.98 and 0.99, respectively. We demonstrate that read count and allelic ratio-based models can achieve a high accuracy (up to 0.98) for detecting fetal trisomy even if the fetal fraction is as low as 2%. Currently, existing commercial NIPT analysis requires at least 4% of fetal fraction, which can be possibly a challenge in case of early gestational age (35 kg/m2). More accurate detection can be achieved at higher sequencing depth using HMM in conjunction with supplemental models, which significantly improve the trisomy detection especially in borderline scenarios (e.g., very low fetal fraction) and enables to perform NIPT even earlier than 10 weeks of pregnancy.
  • Gordevičius, Juozas; Narmontė, Milda; Gibas, Povilas; Kvederavičiūtė, Kotryna; Tomkutė, Vita; Paluoja, Priit; Krjutškov, Kaarel; Salumets, Andres; Kriukienė, Edita (2020)
    BackgroundMassively parallel sequencing of maternal cell-free DNA (cfDNA) is widely used to test fetal genetic abnormalities in non-invasive prenatal testing (NIPT). However, sequencing-based approaches are still of high cost. Building upon previous knowledge that placenta, the main source of fetal circulating DNA, is hypomethylated in comparison to maternal tissue counterparts of cfDNA, we propose that targeting either unmodified or 5-hydroxymethylated CG sites specifically enriches fetal genetic material and reduces numbers of required analytical sequencing reads thereby decreasing cost of a test.MethodsWe employed uTOPseq and hmTOP-seq approaches which combine covalent derivatization of unmodified or hydroxymethylated CG sites, respectively, with next generation sequencing, or quantitative real-time PCR.ResultsWe detected increased 5-hydroxymethylcytosine (5hmC) levels in fetal chorionic villi (CV) tissue samples as compared with peripheral blood. Using our previously developed uTOP-seq and hmTOP-seq approaches we obtained whole-genome uCG and 5hmCG maps of 10 CV tissue and 38 cfDNA samples in total. Our results indicated that, in contrast to conventional whole genome sequencing, such epigenomic analysis highly specifically enriches fetal DNA fragments from maternal cfDNA. While both our approaches yielded 100% accuracy in detecting Down syndrome in fetuses, hmTOP-seq maintained such accuracy at ultra-low sequencing depths using only one million reads. We identified 2164 and 1589 placenta-specific differentially modified and 5-hydroxymethylated regions, respectively, in chromosome 21, as well as 3490 and 2002 Down syndrome-specific differentially modified and 5-hydroxymethylated regions, respectively, that can be used as biomarkers for identification of Down syndrome or other epigenetic diseases of a fetus.ConclusionsuTOP-seq and hmTOP-seq approaches provide a cost-efficient and sensitive epigenetic analysis of fetal abnormalities in maternal cfDNA. The results demonstrated that T21 fetuses contain a perturbed epigenome and also indicated that fetal cfDNA might originate from fetal tissues other than placental chorionic villi. Robust covalent derivatization followed by targeted analysis of fetal DNA by sequencing or qPCR presents an attractive strategy that could help achieve superior sensitivity and specificity in prenatal diagnostics.
  • Sauk, Martin; Zilina, Olga; Kurg, Ants; Ustav, Eva-Liina; Peters, Maire; Paluoja, Priit; Roost, Anne Mari; Teder, Hindrek; Palta, Priit; Brison, Nathalie; Vermeesch, Joris R.; Krjutskov, Kaarel; Salumets, Andres; Kaplinski, Lauris (2018)
    Non-invasive prenatal testing (NIPT) is a recent and rapidly evolving method for detecting genetic lesions, such as aneuploidies, of a fetus. However, there is a need for faster and cheaper laboratory and analysis methods to make NIPT more widely accessible. We have developed a novel software package for detection of fetal aneuploidies from next-generation low-coverage whole genome sequencing data. Our tool - NIPTmer - is based on counting pre-defined per-chromosome sets of unique k-mers from raw sequencing data, and applying linear regression model on the counts. Additionally, the filtering process used for k-mer list creation allows one to take into account the genetic variance in a specific sample, thus reducing the source of uncertainty. The processing time of one sample is less than 10 CPU-minutes on a high-end workstation. NIPTmer was validated on a cohort of 583 NIPT samples and it correctly predicted 37 non-mosaic fetal aneuploidies. NIPTmer has the potential to reduce significantly the time and complexity of NIPT post-sequencing analysis compared to mapping-based methods. For non-commercial users the software package is freely available at
  • Lassmann-Klee, Paul; Piirilä, Päivi L.; Brumpton, Ben; Larsson, Matz; Sundblad, Britt-Marie; Põlluste, Jaak; Juusela, Maria; Rouhos, Annamari; Meren, Mari; Lindqvist, Ari; Kankaanranta, Hannu; Backman, Helena; Langhammer, Arnulf; Ronmark, Eva; Lundbäck, Bo; Sovijärvi, Anssi (2020)
    The prevalence of asthma is higher in Sweden and Finland than in neighbouring eastern countries including Estonia. Corresponding difference in bronchial eosinophilic inflammation could be studied by FENO measurements. We aimed to compare FENO in adult general populations of Sweden, Finland, and Estonia, to test the plausibility of the west-east disparity hypothesis of allergic diseases. We conducted clinical interviews (N = 2658) with participants randomly selected from the general populations in Sweden (Stockholm and Örebro), Finland (Helsinki), and Estonia (Narva and Saaremaa), and performed FENO (n = 1498) and skin prick tests (SPT) in 1997–2003. The median (interquartile range) of FENO (ppb) was 15.5 (9.3) in Sweden, 15.4 (13.6) in Finland and 12.5 (9.6) in Estonia. We found the lowest median FENO values in the Estonian centres Saaremaa 13.1 (9.5) and Narva 11.8 (8.6). In the pooled population, asthma was associated with FENO ≥25 ppb, odds ratio (OR) 3.91 (95% confidence intervals: 2.29–6.32) after adjusting for SPT result, smoking, gender and study centre. A positive SPT test increased the likelihood of asthma OR 3.19 (2.02–5.11). Compared to Saaremaa, the likelihood of having asthma was higher in Helsinki OR 2.40 (1.04–6.02), Narva OR 2.45 (1.05–6.19), Örebro OR 3.38 (1.59–8.09), and Stockholm OR 5.54 (2.18–14.79). There was a higher prevalence of asthma and allergic airway inflammation in adult general populations of Sweden and Finland compared to those of Estonia. Atopy and elevated FENO level were independently associated with an increased risk of asthma. In conclusion, the findings support the earlier west-east disparity hypothesis of allergic diseases.
  • Zou, Xiaochen; Mottus, Matti; Tammeorg, Priit; Lizarazo Torres, Clara; Takala, Tuure; Pisek, Jan; Makela, Pirjo; Stoddard, F. L.; Pellikka, Petri (2014)
  • Heinilä, Anna Maaria Kirsikka; Salminen, Miia; Metsämäki, Sari; Pellikka, Petri Kauko Emil; Koponen, Sampsa; Pulliainen, Jouni (2019)
    We aim a better understanding of the effect of spring-time snow melt on the remotely sensed scene reflectance by using an extensive amount of optical spectral data obtained from an airborne hyperspectral campaign in Northern Finland. We investigate the behaviour of thin snow reflectance for different land cover types, such as open areas, boreal forests and treeless fells. Our results not only confirm the generally known fact that the reflectance of a melting thin snow layer is considerably lower than that of a thick snow layer, but we also present analyses of the reflectance variation over different land covers and in boreal forests as a function of canopy coverage. According to common knowledge, the highly variating reflectance spectra of partially transparent, most likely also contaminated thin snow pack weakens the performance of snow detection algorithms, in particular in the mapping of Fractional Snow Cover (FSC) during the end of the melting period. The obtained results directly support further development of the SCAmod algorithm for FSC retrieval, and can be likewise applied to develop other algorithms for optical satellite data (e.g. spectral unmixing methods), and to perform accuracy assessments for snow detection algorithms. A useful part of this work is the investigation of the competence of Normalized Difference Snow Index (NDSI) in snow detection in late spring, since it is widely used in snow mapping. We conclude, based on the spectral data analysis, that the NDSI-based snow mapping is more accurate in open areas than in forests. However, at the very end of the snow melting period the behavior of the NDSI becomes more unstable and unpredictable in non-forests with shallow snow, increasing the inaccuracy also in non-forested areas. For instance in peatbogs covered by melting snow layer (snow depth <30 cm) the mean NDSI-0.6 was observed, having coefficient of variation as high as 70%, whereas for deeper snow packs the mean NDSI shows positive values.
  • Lu, Peng; Leppäranta, Matti; Cheng, Bin; Li, Zhijun; Istomina, Larysa; Heygster, Georg (2018)
    Pond color, which creates the visual appearance of melt ponds on Arctic sea ice in summer, is quantitatively investigated using a two-stream radiative transfer model for ponded sea ice. The upwelling irradiance from the pond surface is determined and then its spectrum is transformed into RGB (red, green, blue) color space using a colorimetric method. The dependence of pond color on various factors such as water and ice properties and incident solar radiation is investigated. The results reveal that increasing underlying ice thickness H-i enhances both the green and blue intensities of pond color, whereas the red intensity is mostly sensitive to H-i for thin ice (H-i <1.5 m) and to pond depth H-p for thick ice (H-i > 1.5 m), similar to the behavior of meltpond albedo. The distribution of the incident solar spectrum F-0 with wavelength affects the pond color rather than its intensity. The pond color changes from dark blue to brighter blue with increasing scattering in ice, and the influence of absorption in ice on pond color is limited. The pond color reproduced by the model agrees with field observations for Arctic sea ice in summer, which supports the validity of this study. More importantly, the pond color has been confirmed to contain information about meltwater and underlying ice, and therefore it can be used as an index to retrieve H-i and H-p. Retrievals of H-i for thin ice (H-i <1 m) agree better with field measurements than retrievals for thick ice, but those of H-p are not good. The analysis of pond color is a new potential method to obtain thin ice thickness in summer, although more validation data and improvements to the radiative transfer model will be needed in future.
  • Majasalmi, Titta; Rautiainen, Miina; Stenberg, Pauline; Manninen, Terhikki (2015)
    Remote sensing of the fraction of absorbed Photosynthetically Active Radiation (fPAR) has become a timely option to monitor forest productivity. However, only a few studies have had ground reference fPAR datasets containing both forest canopy and understory fPAR from boreal forests for the validation of satellite products. The aim of this paper was to assess the performance of two currently available satellite-based fPAR products: MODIS fPAR (MOD15A2, C5) and GEOV1 fPAR (g2_BIOPAR_FAPAR), as well as an NDVI-fPAR relationship applied to the MODIS surface reflectance product and a Landsat 8 image, in a boreal forest site in Finland. Our study area covered 16 km(2) and field data were collected from 307 forest plots. For all plots, we obtained both forest canopy fPAR and understory fPAR. The ground reference total fPAR agreed better with GEOV1 fPAR than with MODIS fPAR, which showed much more temporal variation during the peak-season than GEOV1 fPAR. At the chosen intercomparison date in peak growing season, MODIS NDVI based fPAR estimates were similar to GEOV1 fPAR, and produced on average 0.01 fPAR units smaller fPAR estimates than ground reference total fPAR. MODIS fPAR and Landsat 8 NDVI based fPAR estimates were similar to forest canopy fPAR.