Tiago Loureiro

Boosting the Detection of Transposable Elements Using Machine Learning

Transposable Elements (TE) are sequences of DNA that move and transpose within a genome. TEs, as mutation agents, are quite important for their role in both genome alteration diseases and on species evolution. Several tools have been developed to discover and annotate TEs but no single one achieves good results on all different types of TEs. In this paper we evaluate the performance of several TEs detection and annotation tools and investigate if Machine Learning techniques can be used to improve their overall detection accuracy. The results of an in silico evaluation of TEs detection and annotation tools indicate that their performance can be improved by using machine learning classifiers.

Digital ulcers in systemic sclerosis: role of flow-mediated dilatation and capillaroscopy as risk assessment tools

AimThe aim of this study was to evaluate macrovascular endothelial dysfunction and microvascular damage as clinical markers of peripheral microangiopathy in patients with Raynaud’s phenomenon (RP).Patients and methodsSeventy-seven secondary RP with systemic sclerosis, 32 primary RP and 34 healthy controls were included in our study. Secondary RP patients were divided into two subgroups: 39 with digital ulcers (DU) and 38 without digital ulcers (non-DU).ResultsPatients with DU had significantly lower flow-mediated dilatation values (5.34 ± 7.49%) compared to non-DU patients (16.21 ± 11.31%), primary RP (17.96 ± 12.78%) and controls (20.17 ± 8.86%), p<0.001, favouring macrovascular endothelium dysfunction. Regarding microvascular damage, the DU group had a predominately capillaroscopic late pattern (71.1%) whereas non-DU patients had an active pattern (56.4%). The microangiopathy evolution score was significantly higher in the DU group compared to the non-DU group (4.79 ± 1.82 vs. 1.79 ± 1.56, p<0.001). Flow-mediated dilation was significantly lower in late pattern (6.13 ± 7.09%) compared to active (12.58 ± 10.66%) and early patterns (17.72 ± 14.90%), p = 0.016 and p = 0.044 respectively.ConclusionsLow flow-mediated dilatation and microvascular damage in capillaroscopy are early clinical markers of DU risk in RP patients.

Improving the performance of Transposable Elements detection tools.

Transposable Elements (TE) are sequences of DNA that move and transpose within a genome. TEs, as mutation agents, are quite important for their role in both genome alteration diseases and on species evolution. Several tools have been developed to discover and annotate TEs but no single tool achieves good results on all different types of TEs. In this paper we evaluate the performance of several TEs detection and annotation tools and investigate if Machine Learning techniques can be used to improve their overall detection accuracy. The results of an in silico evaluation of TEs detection and annotation tools indicate that their performance can be improved by using machine learning constructed classifiers.

Iatrogenic Complications Following Laser Ablation of Varicose Veins

During the past decade, Endovenous Laser Ablation (EVLA) has been introduced as a minimally invasive alternative to high ligation and open surgical stripping of the incompetent Great Saphenous Vein (GSV) or Short Saphenous Vein (SSV). There is great variability in EVLA protocols and at the present time there is data from more than 60 publications with more than 15000 EVLA treatments reporting good clinical results but also undesired side effects. Post-operative pain in the operated area is rated from slight to moderate by most patients (81, 5%),post-operative induration of the truncal vein can be expected in 78, 1% of patients and perivenous ecchymosis and hematoma are observed in an average of 52% of patients. Further complications such as persistent dysesthesia after nerve lesions (0, 8%) and burns on the skin (0, 2%) are reported in a minority of cases. Great care must be taken to ensure adequate tumescent anesthesia and light dosimetry in order to avoid post-operative paresthesia.