I study computational and quantitative biology with a focus on network aging. This site is to serve as my note-book and to effectively communicate with my students and collaborators. Every now and then, a blog may be of interests to other researchers or teachers. Views in this blog are my own. All rights of research results and findings on this blog are reserved. See also http://youtube.com/c/hongqin
Using cached https://files.pythonhosted.org/packages/03/ad/d732a5d9d50bfcd8aeb6e4a266065a8868829388e4e2b529ff689f1fc923/tensorflow-1.8.0-cp36-cp36m-macosx_10_11_x86_64.whl
Requirement not upgraded as not directly required: wheel>=0.26 in ./lib/python3.6/site-packages (from tensorflow) (0.31.1)
Collecting protobuf>=3.4.0 (from tensorflow)
Using cached https://files.pythonhosted.org/packages/6d/7e/51c91b28cb8446ebd7231d375a2025bca4c59d15ddc0cf2dd0867b400cd7/protobuf-3.6.0-cp36-cp36m-macosx_10_6_intel.macosx_10_9_intel.macosx_10_9_x86_64.macosx_10_10_intel.macosx_10_10_x86_64.whl
Requirement not upgraded as not directly required: six>=1.10.0 in /anaconda3/lib/python3.6/site-packages (from tensorflow) (1.11.0)
Collecting termcolor>=1.1.0 (from tensorflow)
Collecting gast>=0.2.0 (from tensorflow)
Collecting absl-py>=0.1.6 (from tensorflow)
Collecting astor>=0.6.0 (from tensorflow)
Using cached https://files.pythonhosted.org/packages/b2/91/cc9805f1ff7b49f620136b3a7ca26f6a1be2ed424606804b0fbcf499f712/astor-0.6.2-py2.py3-none-any.whl
/anaconda3/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
from ._conv import register_converters as _register_converters
2018-06-29 21:22:49.100997: W tensorflow/core/framework/op_def_util.cc:346] Op BatchNormWithGlobalNormalization is deprecated. It will cease to work in GraphDef version 9. Use tf.nn.batch_normalization().
2018-06-29 21:22:49.315593: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
A Survey on Deep Learning in Medical Image Analysis
Geert Litjens, Thijs Kooi, Babak Ehteshami Bejnordi, Arnaud Arindra Adiyoso Setio, Francesco Ciompi,
Mohsen Ghafoorian, Jeroen A.W.M. van der Laak, Bram van Ginneken, Clara I. Sa ́nchez
Diagnostic Image Analysis Group
Radboud University Medical Center
Nijmegen, The Netherlands
"Currently, the most popular models are trained end-
to-end in a supervised fashion, greatly simplifying
the training process. The most popular architectures
are convolutional neural networks (CNNs) and recur-
rent neural networks (RNNs). CNNs are currently
most widely used in (medical) image analysis, although
RNNs are gaining popularity. "
The second key difference between CNNs and MLPs,
is the typical incorporation of pooling layers in CNNs,
where pixel values of neighborhoods are aggregated using a permutation invariant function, typically the max
or mean operation. This induces a certain amount of
translation invariance and again reduces the amount of
parameters in the network. At the end of the convo-
lutional stream of the network, fully-connected layers
(i.e. regular neural network layers) are usually added,
where weights are no longer shared. Similar to MLPs,
a distribution over classes is generated by feeding the
activations in the final layer through a softmax function
and the network is trained using maximum likelihood.
Adjacency matrix in Yuan exact controllability paper use column_node -> row_node, which is a mirror of the conventional row_node -> column_node. These mirrored adjacency matrix for direct graphs do not change controllability analysis using the matrix based method, it seems to me.
Fall 2018 courses have been created, and are now available to you in UTC Learn. Students will be added to courses one (1) week before the first day of classes (8/13/2018). If you would like to merge any of your courses, please complete the course merge request form located at:
Nano SIM is both smaller and approximately 15% thinner than the earlier Micro SIM(3FF) standard as well as the Mini SIM (2FF) cards that were ubiquitous for many years and people commonly refer to simply as SIM cards.Apr 9, 2018 Nano SIM is the fourth version, or the "fourth form factor" (4FF) of the SIM standard and measures a mere 12.3 mm by 8.8 mm by 0.67 mm, but still holds the same amount of data as earlier SIM cards.Apr 9, 2018
Fully functional CRISPR/Cas enzymes will introduce a double-strand break (DSB) at a specific location based on a gRNA-defined target sequence. DSBs are preferentially repaired in the cell by non-homologous end joining (NHEJ), a mechanism which frequently causes insertions or deletions (indels) in the DNA. Indels often lead to frameshifts, creating loss of function alleles.
To introduce specific genomic changes, researchers use ssDNA or dsDNA repair templates with homology to the DNA flanking the DSB and a specific edit close to the gRNA PAM site. When a repair template is present, the cell may repair a DSB using homology-directed repair (HDR) instead of NHEJ. In most experimental systems, HDR occurs at a much lower efficiency than NHEJ.