1. |
|
|
2. |
- Andersson, Björn, 1977, et al.
(author)
-
Development of a machine learning framework for radiation biomarker discovery and absorbed dose prediction.
- 2023
-
In: Frontiers in oncology. - 2234-943X. ; 13
-
Journal article (peer-reviewed)abstract
- Molecular radiation biomarkers are an emerging tool in radiation research with applications for cancer radiotherapy, radiation risk assessment, and even human space travel. However, biomarker screening in genome-wide expression datasets using conventional tools is time-consuming and underlies analyst (human) bias. Machine Learning (ML) methods can improve the sensitivity and specificity of biomarker identification, increase analytical speed, and avoid multicollinearity and human bias.To develop a resource-efficient ML framework for radiation biomarker discovery using gene expression data from irradiated normal tissues. Further, to identify biomarker panels predicting radiation dose with tissue specificity.A strategic search in the Gene Expression Omnibus database identified a transcriptomic dataset (GSE44762) for normal tissues radiation responses (murine kidney cortex and medulla) suited for biomarker discovery using an ML approach. The dataset was pre-processed in R and separated into train and test data subsets. High computational cost of Genetic Algorithm/k-Nearest Neighbor (GA/KNN) mandated optimization and 13 ML models were tested using the caret package in R. Biomarker performance was evaluated and visualized via Principal Component Analysis (PCA) and dose regression. The novelty of ML-identified biomarker panels was evaluated by literature search.Caret-based feature selection and ML methods vastly improved processing time over the GA approach. The KNN method yielded overall best performance values on train and test data and was implemented into the framework. The top-ranking genes were Cdkn1a, Gria3, Mdm2 and Plk2 in cortex, and Brf2, Ccng1, Cdkn1a, Ddit4l, and Gria3 in medulla. These candidates successfully categorized dose groups and tissues in PCA. Regression analysis showed that correlation between predicted and true dose was high with R2 of 0.97 and 0.99 for cortex and medulla, respectively.The caret framework is a powerful tool for radiation biomarker discovery optimizing performance with resource-efficiency for broad implementation in the field. The KNN-based approach identified Brf2, Ddit4l, and Gria3 mRNA as novel candidates that have been uncharacterized as radiation biomarkers to date. The biomarker panel showed good performance in dose and tissue separation and dose regression. Further training with larger cohorts is warranted to improve accuracy, especially for lower doses.
|
|
3. |
- Andersson, Charlotte, et al.
(author)
-
Biodistribution of I-131 in mice is influenced by circadian variations
- 2020
-
In: Scientific Reports. - : Springer Science and Business Media LLC. - 2045-2322. ; 10:1
-
Journal article (peer-reviewed)abstract
- Effects of radiation and biodistribution of radionuclides are often studied in animal models. Circadian rhythm affects many biological functions and may influence the biokinetics of radionuclides and observed responses. The aim of this study was to investigate if the time during the day of I-131 injection affects the biodistribution and absorbed dose to tissues in mice. Biodistribution studies were conducted on male C57BL/6 N mice for three diurnal time-series: the animals were i.v. injected with 160 kBq I-131 at 8 am, 12 pm or 4 pm. The activity concentration in organs and tissues was measured at 1 h to 7 days after administration and absorbed dose at day 7 was determined. Comparison between the three time-series showed statistically significant differences in activity concentration in all investigated tissues and organs. Administration performed at 12 pm resulted in general in higher absorbed dose to the organs than injection performed at 8 am and 4 pm. Time of day of administration affects the biodistribution of I-131 in mice and consequently the absorbed dose to individual organs. These findings advocate that subsequent biodistribution studies and dosimetry calculations should consider time-point of administration as a variable that could influence the results.
|
|
4. |
-
Friluftsliv explored : An environmental and outdoor teaching approach for knowledge, emotions and quality of life
- 2021
-
Editorial collection (other academic/artistic)abstract
- Friluftsliv explored doesn’t only include nature knowledge, techniques in the outdoors and outdoor pedagogics but also covers ecology, human ecology, geography, environmental and societal questions, history, health, biology, craft and lots of practical activities -both for urban and rural friluftsliv. In this translation to English of the revised fifth edition of the Swedish book there are many activities and the text is suitable for the modern day.Friluftsliv embraces the feeling around the campfire, paddling along winding rivers and walking towards the distant blue mountains. But, it is also to whittle a stick, to remember your waterproofs and to find your way home.Knowledge emerges when you combine imagination with facts and the glint in your eyes, using all our outdoor environments: forests, water, the coast, mountains and the nature close at hand.Emotion is to swim in crystal clear water far out in the archipelago and to see the clouds gliding across the sky. But also, to be able to present other sides of yourself, to be fascinated by your own body, the struggling ant and the sight of frost on trees.Quality of life is to experience friluftsliv – as it happens!
|
|
5. |
- Lodin, Karin, et al.
(author)
-
Patient global assessment and inflammatory markers in patients with idiopathic inflammatory myopathies : A longitudinal study
- 2024
-
In: Seminars in Arthritis & Rheumatism. - : Elsevier. - 0049-0172 .- 1532-866X. ; 65
-
Journal article (peer-reviewed)abstract
- AimTo explore if patient global assessment (PGA) is associated with inflammation over time and if associations are explained by other measures of disease activity and function in patients with idiopathic inflammatory myopathies (IIM).MethodsPGA and systemic inflammatory markers prospectively collected over five years were retrieved from the International MyoNet registry for 1200 patients with IIM. Associations between PGA, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) and creatine kinase (CK) were analyzed using mixed models. Mediation analysis was used to test if the association between PGA and inflammatory markers during the first year of observation could be explained by measures of disease activity and function.ResultsPGA improved, and inflammatory markers decreased during the first year of observation. In the mixed models, high levels of inflammatory markers were associated with worse PGA in both men and women across time points during five years of observation. In men, but not in women, the association between elevated ESR, CRP and poorer PGA was explained by measures of function and disease activity. With a few exceptions, the association between improved PGA and reduced inflammatory markers was partially mediated by improvements in all measures of function and disease activity.ConclusionIncreased levels of systemic inflammation are associated with poorer PGA in patients with IIM. In addition to known benefits of lowered inflammation, these findings emphasize the need to reduce systemic inflammation to improve subjective health in patients with IIM. Furthermore, the results demonstrate the importance of incorporating PGA as an outcome measure in clinical practice and clinical trials.
|
|
6. |
- Lötstedt, Britta
(author)
-
Spatial mapping of bacteria and transcriptomes
- 2022
-
Doctoral thesis (other academic/artistic)abstract
- Novel insights into biological functions and mechanisms, cell networks and evolutionary relationships are gained through development of sequencing technologies and sequencing based applications. Massively parallel sequencing has enabled analysis of big data at gene and protein expression levels, but has also characterized bacterial communities. Additionally, different technological advancements enabled us to track those expression changes in single cells, to reveal insights into rare cell populations, or with added spatial resolution, to explore highly complex environments such as tissues. This thesis gives an overview of different technical, biological and computational methods used in genomics today with a specific focus on spatial techniques for detailed tissue characterization. This is followed by a chapter summarizing recent scientific contributions made by the author that have been included as part of this thesis. In Paper I, 16S sequencing was used to study the diversity and composition of bacterial communities with specific focus on the aerodigestive microbiome in children who had undergone a lung transplant. Potential connections between the microbiome and irregular gastric muscle movements were also examined. Patients with a lung transplant had significantly lower microbial diversity in the gastric and oropharyngeal sites as compared to controls, however, lung transplant recipients showed similar bacterial compositions, independent of motility status. Samples in the lung transplant patient group were in general dominated by Staphylococcaceae but Streptococcus, Prevotella and Veillonella were common in the gastric and oropharyngeal samples. Next, an automated method for simultaneous spatial analysis of both gene and antibodybased protein expression in tissue sections, named SM-Omics, was developed in Paper II. SM-Omics enabled simultaneous detection of proteins, by using either immunofluorescence or DNAbarcoded antibodies, and analysis of the spatial transcriptome in the same tissue section. SM-Omics was applied to the mouse brain and spleen and obtained correlated spatial patterns between respective gene and antibody measurements. The method allowed processing of up to 64 in situ spatial reactions or up to 96 sequencing-ready libraries, of high complexity, in a ~2 days process. The spatial host-microbiome sequencing method, presented in Paper III, was used to concurrently study the spatial environment created between bacteria and host cells within a tissue section. Using spatial host-microbiome sequencing, colonic sections from three different mouse models were examined by simultaneous in situ capture of both mRNA and 16S sequences, followed by sequencing and taxonomic assignment of bacterial 16S sequences using a deep learning model. ~17,000 genes and 39 bacteria genera across 16 different morphological regions were quantitatively assessed in the mouse colon. We reported specific genera in the interfold and lumen regions of the colon, as well as spatially variable genes across 100 tissue sections. To better understand genotype-relevant changes impacted by bacterial presence, we defined cell-type specific interactions described with sets of activated pathways. Finally, consecutive tissue sections of multiple synovial biopsies from patients suffering from rheumatoid arthritis were processed using the Spatial Transcriptomics method and sequenced in Paper IV. The alignment and transformation of the consecutive tissue sections enabled spatial profiling in 3D of genes and cell types within the biopsies. Spatially variable gene expression patterns revealed clusters radially distributed around organized structures of infiltrating leukocytes (TLOs). In patients with developed TLOs, these structures contained proinflammatory B cells, while the surrounding areas were high in fibroblasts.
|
|
7. |
|
|