Wednesday, February 27, 2019

Engineering college facts

freshmen/transfer, enrollmet at 269 in fall 2018, transfer enrollment fluctuate,

graudate student. flat for cs phd

undergraudate graduation, increasing,

as high as 40%
DFW rates by courses: CHEM1110, 1120, 3010, 3020, 3710
Physic 2310
Math 1710, 1720, 1730, 1830, 1920, 1950, 1960, 2100, 2450, 2550
math camp in summer was proposed

nonwhite 26% at college of engineering

femal 16% below national average 18%?

20% of UTC funding from Engineering,

$4.5M in FY2018, 0.7M of CS

MS engin manag 7th in nation

student credit hours 23K in 2018, 21,222 in 2017

Cleveland stat, cha s U, Dalon, Covrnent, Roane state, Motlow S U

7 new faculty hires

5100 sq ft resaerch space, lab and work space, construction starts in summer 2018

relieft from diff tution obligation?

April 18, tech sympo

Tuesday, February 26, 2019

iCompBIO REU sites

online applicationform 


Sunday, February 24, 2019

python deep learning notes

stochastic gradient descent is so called because only batch of data are used each time.
========gradient descent
# Set the learning rate: learning_rate
learning_rate = 0.01

# Calculate the predictions: preds
preds = (weights * input_data).sum()

# Calculate the error: error
error = preds - target

# Calculate the slope: slope
slope = 2 * input_data * error

# Update the weights: weights_updated
weights_updated = weights - learning_rate * slope

# Get updated predictions: preds_updated
preds_updated = (weights_updated * input_data).sum()

# Calculate updated error: error_updated
error_updated = preds_updated - target

# Print the original error

# Print the updated error


# Import necessary modules
import keras
from keras.layers import Dense
from keras.models import Sequential

# Specify the model
n_cols = predictors.shape[1]
model = Sequential()
model.add(Dense(50, activation='relu', input_shape = (n_cols,)))
model.add(Dense(32, activation='relu'))

# Compile the model
model.compile(optimizer='adam', loss='mean_squared_error')

# Verify that model contains information from compiling
print("Loss function: " + model.loss)

# Fit the model, target)

# Import necessary modules
import keras
from keras.layers import Dense
from keras.models import Sequential
from keras.utils import to_categorical

# Convert the target to categorical: target
target = to_categorical(df.survived)

# Set up the model
model = Sequential()

# Add the first layer
model.add(Dense(32, activation='relu', input_shape=(n_cols,)))

# Add the output layer
model.add(Dense(2, activation='softmax'))

# Compile the model

# Fit the model, target)

dead neuron versus vanishing gradient


Thursday, February 21, 2019

FreightTech Innovation Challenge: A 24-Hour Transportation and Logistics Use Case Competition,

I’m excited to announce the inaugural FreightTech Innovation Challenge: A 24-Hour Transportation and Logistics Use Case Competition, taking place March 29–30, 2019 in Chattanooga, TN. Presented in collaboration with FreightWaves and CO.LAB, this event is a 24-hour competition for college students from across the nation to team up and take on some of the greatest challenges in transportation and logistics.

From now until March 8, we invite students who are interested in business, supply chain, technology, computer science, data, and logistics (and more) to apply for this immersive experience that could ultimately impact their career paths. Participating students will work directly with major industry players and experts to find solutions to real challenges in the industry, with a chance to win cash prizes and find potential future employers.

Here are the benefits for participating students:
·  Opportunity to work closely with a team of student peers and industry experts to solve unique industry challenges
·  Network with fast-growing companies and find potential future employers
·  Chance to win a cash prize and gain recognition for yourself and your university (cash prizes awarded to top three teams: 1st Place - $5,000; 2nd Place - $3,000; 3rd Place - $1,000)
·  Weekend to discover Chattanooga, TN—a leading hub for the transportation industry—also known as “Freight Alley” (participating teams will receive housing vouchers)

We ask that all students interested apply at, where they can indicate whether they are part of a team or applying as an individual. 

You can find more information at or on the information sheet attached to this email. Included on the information sheet is the schedule (may be subject to change) that outlines the activities for the weekend. Please let me know if you have any questions.

Why Chattanooga?

Chattanooga—known as the Scenic City due to its beauty and outdoor recreational scene—has grown to become athriving new hub for startups and large companies alike. With a supportive ecosystem for businesses, Chattanooga has also established itself as the heart of “Freight Alley,” with an increasingly growing number of companies in the transportation and logistics industry due to its position as the epicenter of freight traffic in the Southeast.

“If you start a company there to serve the trucking industry, you have more expertise about what the needs are, and more customers and partners there in Chattanooga as opposed to New York City, Boston and San Francisco. Chattanooga is the “Silicon Valley” of Trucking”
— Steve Case, AOL Co-Founder and Co-Founder of Rise of the Rest

-20 freezers, fisher isothermal

Monday, February 18, 2019

*** used lab equipment

new life science

Saturday, February 16, 2019

Systems biology of aging

Why individuals with similar genetic makeup live to different lifespans? How can we extend health lifespan? Our group approach these questions through mathematical formulation and computational analysis of genomics data to generate predictive models based on gene/protein networks. Our group also develop various machine learning tools to extrapolate useful information from biological big data.

Monday, February 11, 2019

Free evaluation

SALG URSSA .(BIO funded sites use this free evaluations)

Wednesday, February 6, 2019

gaussian and laplacian filters

from wikipedia
Gaussian smoothing is commonly used with edge detection. Most edge-detection algorithms are sensitive to noise; the 2-D Laplacian filter, built from a discretization of the Laplace operator, is highly sensitive to noisy environments. Using a Gaussian Blur filter before edge detection aims to reduce the level of noise in the image, which improves the result of the following edge-detection algorithm. This approach is commonly referred to as Laplacian of Gaussian, or LoG filtering.

Tuesday, February 5, 2019

data science programming textbook

python for data analysis. Oreily or Packt.

Python for biologist

advanced python for biologist

python programming for biology:Bioinformatics and Beyond 1st Edition

Bioinformatics with Python Cookbook: Learn how to use modern Python bioinformatics libraries and applications to do cutting-edge research in computational biology, 2nd Edition Paperback

Friday, February 1, 2019

*** desktop flow cytometer and cell counter survey

LUNA-FL . $9.7K 

Moxi Go 2, VWR $21K

Guava Millipore

CytoFlex $76K ~ $110K. 

advice from others:
"You might also look into the LUNA-FL or Countess FLII systems if the end use is as described, they are basically automated low-power epifluorescent scopes.  There are disposable and reusable slide formats. These are instruments much cheaper in comparison.  However, the Countess might not be able to assess yeast viability.

I have a Countess that we use for simple stuff, and we can get rough phenotypes for abundant surface markers.

I cannot recommend the Muse because it is my understanding that it’s completely kit based.  The Guava has been reliable in my experience, but I haven’t used one in a few years. "

SpectraMax® MiniMax™ 300 Imaging Cytometer, Molecular Devices, $45K 

"I strongly recommend trying your yeast samples on a demo Countess II/II FL instrument before purchasing one.  Yeast cells are generally very small and may be below the size cut-off for accurate counting and viability.  The instrument will see and count any cells larger than 4 um in diameter, but will only give correct viability for cells larger than 7 um.  If you open the histogram option on the Countess II results screen, you can see the diameter range the instrument sees for the sample. Unfortunately, we do not have example data for yeast.  "

Main fluophores: 
  Propidum Iodide (PI), dihydroethydium (DHE), and dihydrorhodamine (DHR)

My understanding of the basic configuration are (based on CALIBUR):
 blue laser 488nm + red laser 635nm (optional)
Detector channels: FSC, SSC, FL1 green 530nm, FL2 yellow 585, FL3 red 670nm, with an optional  FL4 661nm.

Vendors and models

BioRad S3e Cell Sorter


ACEA NovoCyte 1000 $40K ?
Novote Cyte model 2000, 3000 have more options. CytoFlex

Accuri C6PLUS  $49K

EmdMillipore Muse: $15K,nav#mch100102
Con: Closed system. Cannot be modified ?

Attune NxT Flow Cytomer

Stratedigm FLow Cytometer

Sales rep from BD on Gallios and CytoFlex
As per my voice message, I understand you are looking for a quote for 6 color Gallios Flow Cytometer, but we now have a new, much smaller flow cytometer with greater capability than the Gallios, and less expensive.   It is called the CytoFLEX Flow Cytometer.   The CytoFLEX is available in 21 different configurations from a 1 laser/4 color system to a 3 laser/13 color system in a small box (16in x 16in x 13in).   It is completely upgradable to any configuration at any time.   A 96 well plate loader option is available as are 561 nm and 375 nm laser options.   Independent testing has proven this instrument to be one of the most sensitive instruments in the market today.
 If the 561 or 375 nm lasers interest you, please let me know and I will forward additional information on our CytoFLEX “S” System.