Revisiting the out of Africa event with a deep-learning approach

Author(s):  
Francesco Montinaro ◽  
Vasili Pankratov ◽  
Burak Yelmen ◽  
Luca Pagani ◽  
Mayukh Mondal
2020 ◽  
Author(s):  
Francesco Montinaro ◽  
Vasili Pankratov ◽  
Burak Yelmen ◽  
Luca Pagani ◽  
Mayukh Mondal

AbstractAnatomically modern humans evolved around 300 thousand years ago in Africa1. Modern humans started to appear in the fossil record outside of Africa about 100 thousand years ago though other hominins existed throughout Eurasia much earlier2–4. Recently, several researchers argued in favour of a single out of Africa event for modern humans based on whole-genome sequences analyses5–7. However, the single out of Africa model is in contrast with some of the findings from fossil records, which supports two out of Africa8,9, and uniparental data, which proposes back to Africa movement10,11. Here, we used a novel deep learning approach coupled with Approximate Bayesian Computation and Sequential Monte Carlo to revisit these hypotheses from the whole genome sequence perspective. Our results support the back to Africa model over other alternatives. We estimated that there are two successive splits between Africa and out of African populations happening around 60-80 thousand years ago and separated by 12-13 thousand years. One of the populations resulting from the more recent split has to a large extent replaced the older West African population while the other one has founded the out of Africa populations.


2018 ◽  
Vol 6 (3) ◽  
pp. 122-126
Author(s):  
Mohammed Ibrahim Khan ◽  
◽  
Akansha Singh ◽  
Anand Handa ◽  
◽  
...  

2020 ◽  
Vol 17 (3) ◽  
pp. 299-305 ◽  
Author(s):  
Riaz Ahmad ◽  
Saeeda Naz ◽  
Muhammad Afzal ◽  
Sheikh Rashid ◽  
Marcus Liwicki ◽  
...  

This paper presents a deep learning benchmark on a complex dataset known as KFUPM Handwritten Arabic TexT (KHATT). The KHATT data-set consists of complex patterns of handwritten Arabic text-lines. This paper contributes mainly in three aspects i.e., (1) pre-processing, (2) deep learning based approach, and (3) data-augmentation. The pre-processing step includes pruning of white extra spaces plus de-skewing the skewed text-lines. We deploy a deep learning approach based on Multi-Dimensional Long Short-Term Memory (MDLSTM) networks and Connectionist Temporal Classification (CTC). The MDLSTM has the advantage of scanning the Arabic text-lines in all directions (horizontal and vertical) to cover dots, diacritics, strokes and fine inflammation. The data-augmentation with a deep learning approach proves to achieve better and promising improvement in results by gaining 80.02% Character Recognition (CR) over 75.08% as baseline.


2018 ◽  
Vol 15 (1) ◽  
pp. 6-28 ◽  
Author(s):  
Javier Pérez-Sianes ◽  
Horacio Pérez-Sánchez ◽  
Fernando Díaz

Background: Automated compound testing is currently the de facto standard method for drug screening, but it has not brought the great increase in the number of new drugs that was expected. Computer- aided compounds search, known as Virtual Screening, has shown the benefits to this field as a complement or even alternative to the robotic drug discovery. There are different methods and approaches to address this problem and most of them are often included in one of the main screening strategies. Machine learning, however, has established itself as a virtual screening methodology in its own right and it may grow in popularity with the new trends on artificial intelligence. Objective: This paper will attempt to provide a comprehensive and structured review that collects the most important proposals made so far in this area of research. Particular attention is given to some recent developments carried out in the machine learning field: the deep learning approach, which is pointed out as a future key player in the virtual screening landscape.


Sign in / Sign up

Export Citation Format

Share Document