feldman

class: center, middle

name:opening

## Connectome Coding

<br>

Joshua T. Vogelstein

---
      class: top, left

### human brain > supercomputers

| | human brain | Titan |
      | --- | --- | --- |
      | speed (PFLOPS) | 100 | 10  |
      | memory (TB) | 1000 | 10 |
      | cost (10$^6$ USD) |  1 | 100 |
      | volume (m$^3$) | 0.001 | 1000
      | mass (kg) | 1 | 10$^7$|
      | energy (W) | 10 | 10$^7$|
      | efficiency (PFLOPS/W) | 10 | 10$^{-6}$ |

--

.center[how?]

---

### conjecture

the human brain's unprecendented computational capabilities emerge from its  connectivity structure

--

### opportunity

decipher the principles of neural connectivity to yield a 10$^7$ improvement in computational efficiency

---

### what is required

1. experimental data
      2. computational tools
      3. statistical methods

---

<div id="contentframe" style="position:relative; top: 0px; left: -100px;">
      <iframe width="900" height="600" src="https://www.youtube.com/embed/c-NMfp13Uug?start=118" frameborder="0" allowfullscreen></iframe>
      </div>

---

<div id="contentframe" style="position:relative; top: 0px; left: -100px;">
      <iframe width="900" height="600" src="https://youtube.com/embed/1aVNRZtxeIU" frameborder="0" allowfullscreen></iframe>
      </div>

---

### computational tools

1. storage
      2. visualization
      3. connectome estimation
      4. big machine learning
      5. network stats

---
      class: top, left

### storage

<a href="http://neurodata.io/data/">
      <img src="images/table1.png" STYLE="position:relative; WIDTH:100%;"/>
      </a>

---
      class: top, left

### visualization

<a href="https://viz.boss.neurodata.io/#!{'layers':{'EM25K':{'type':'image'_'source':'boss://https://api.boss.neurodata.io/collman/collman15v2/EM25K?'_'opacity':0.5}_'PSD95_488':{'type':'image'_'source':'boss://https://api.boss.neurodata.io/collman/collman15v2/PSD95_488?'_'opacity':0.5_'color':6_'visible':false}_'Synapsin647':{'type':'image'_'source':'boss://https://api.boss.neurodata.io/collman/collman15v2/Synapsin647?'_'opacity':0.5_'visible':false}_'VGluT1_647':{'type':'image'_'source':'boss://https://api.boss.neurodata.io/collman/collman15v2/VGluT1_647?'_'opacity':0.5_'color':1_'visible':false}_'annotation':{'type':'segmentation'_'source':'boss://https://api.boss.neurodata.io/collman/collman15v2/annotation?'}}_'navigation':{'pose':{'position':{'voxelSize':[2.240000009536743_2.240000009536743_70]_'voxelCoordinates':[2335.03564453125_2079.428466796875_21]}}_'zoomFactor':10}}"/>
      <img src="images/AT.png" STYLE="position:relative; WIDTH:100%;"/>
      </a>

---
      class: top, left

### visualization

<a href="https://viz-dev.boss.neurodata.io/#!%7B%27layers%27:%7B%27channel0?window=0,10000%27:%7B%27type%27:%27image%27_%27source%27:%27boss://https://api.boss.neurodata.io/ailey-dev/s3617/channel0?window=0,10000%27_%27opacity%27:0.14_%27color%27:6_%27max%27:0.46%7D_%27channel1?window=0,2000%27:%7B%27type%27:%27image%27_%27source%27:%27boss://https://api.boss.neurodata.io/ailey-dev/s3617/channel1?window=0,2000%27_%27opacity%27:0.65_%27color%27:4_%27max%27:0.75%7D_%27aileydev_ara3_to_s3617_res6%27:%7B%27type%27:%27segmentation%27_%27source%27:%27boss://https://api.boss.neurodata.io/ailey-dev/ara3_to_s3617_res6/aileydev_ara3_to_s3617_res6%27_%27selectedAlpha%27:0.21%7D%7D_%27navigation%27:%7B%27pose%27:%7B%27position%27:%7B%27voxelSize%27:%5B37440_37440_5000%5D_%27voxelCoordinates%27:%5B126.23583221435547_163.23435974121094_529%5D%7D%7D_%27zoomFactor%27:14654.110702108632%7D_%27showScaleBar%27:false%7D"/>
      <img src="images/CLARITY-s3617.png" STYLE="position:relative; WIDTH:70%;"/>
      </a>

---
      class: top, left

### connectome estimation

<a href="http://m2g.io/">
      <img src="images/ndmg-pipeline.png" STYLE="position:relative; WIDTH:100%;"/>
      </a>

---
      class: top, left

### big machine learning

<a href="http://flashx.io/">
      <img src="images/FlashR-perf.png" STYLE="position:absolute; LEFT:100px; WIDTH:350px;"/>
      <img src="images/FlashR-mem.png" STYLE="position:relative; LEFT:400px; WIDTH:350px;"/>
      </a>

---
      class: top, left

### network stats

---
      class:   top, left

## Questions?

<br>
      [jovo@jhu.edu](mailto:jovo at jhu dot edu)
      <br>
      {[bme](http://www.bme.jhu.edu/),[icm](http://icm.jhu.edu/),[cis](http://cis.jhu.edu/),[idies](http://idies.jhu.edu/),[kavli](http://kavlijhu.org/),[cs](http://engineering.jhu.edu/computer-science/), [ams](http://engineering.jhu.edu/ams/), [neuro](http://neuroscience.jhu.edu/)} | [jhu](https://www.jhu.edu/)
      <br>
      <http://brainx.io/feldman>
      <br>

Co-Founder: [NeuroData](http://neurodata.io), [Gigantum](http://gigantum.io)

<img src="images/lion2.JPG" STYLE="position:absolute; TOP:330px; LEFT:100px; HEIGHT:180px;"/>
      <img src="images/world.gif" STYLE="position:absolute; TOP:330px; LEFT:450px; HEIGHT:180px;"/>

<img src="http://brainx.io/images/funding/nsf_fpo.png" STYLE="position:absolute; TOP:550px; LEFT:100px; HEIGHT:80px;"/>

<img src="http://brainx.io/images/funding/nih_fpo.png" STYLE="position:absolute; TOP:550px; LEFT:195px; HEIGHT:80px;"/>

<img src="http://brainx.io/images/funding/iarpa_fpo.jpg" STYLE="position:absolute; TOP:550px; LEFT:300px; HEIGHT:80px;"/>

<img src="http://brainx.io/images/funding/kavli_fpo.png" STYLE="position:absolute; TOP:550px; LEFT:410px; HEIGHT:80px;"/>

<img src="http://brainx.io/images/funding/kndi_fpo.png" STYLE="position:absolute; TOP:550px; LEFT:500px; HEIGHT:80px;"/>

<img src="http://brainx.io/images/funding/darpa_fpo.png" STYLE="position:absolute; TOP:550px; LEFT:700px; HEIGHT:80px;"/>