Cavalcade of Mammals

The CRISPR gene editing system is a major technical advance. It does open up the near term possibility of making a few small changes to a human embryo’s DNA, but I don’t find that particularly interesting or alarming.

What makes CRISPR better than previous tech for gene modification is that it works at high efficiency–1% to 60% with very high specificity. I read a recent paper testing CRISPR on human embryos that reported 50% effectiveness. Given a handful of embryos to work with, there is a very good chance of making a single change in one embryo.

We have very little knowledge or technology for making positive changes to animals which is a huge limitation to genetic ‘engineering’. Mostly what is understood are disease causing (or predisposing) genetic variants. So a single change (maybe in a few years, a handful of changes?) can be made to a human embryo. There are other limits to modifying human embryos apart from lack of knowledge. The more time an embryo or human embryonic stem cell is cultured, the more it is manipulated, the greater the chance of something going wrong, and the child being born with problems. This tech is great for manipulating animals in the lab. If many or most of them have the genetic change, great! If some are born with defects, cull them, or breed another generation and use those in experiments (often the first generation has non-genetic defects that breed away). But these are huge problems if you are working on humans, because things that increase the risk of getting a damaged child are not desirable.

Long term (100-1000 years), when increases in understanding of biology make improvements (or significant changes of any sort) in humans possible, I think what we’ll see is that the people with the least concern for child welfare will be the most willing to experiment on them.

The really exciting possibilities CRISPR opens up is in genetic treatment of human disease in the tissues of kids and adults. There is delivery tech (well tested viral vectors, and a host of other methods) that can get CRISPR into a good percentage of cells (10% to 50+%) in many tissues, and once there, CRISPR will edit a good fraction of those cells. For many diseases, fixing a genetic defect in 1%, 10% or 20% of cells is enough to treat the disease, so genetic treatment of host of diseases is now possible. Things like hemophilia, some muscular dystrophy, maybe Huntington’s Disease, metabolic diseases, Parkinson’s disease, and on and on. There will be a lot of exciting advances turning that ‘possible’ into actual treatments for different diseases over the next decade or two.

The other major effect of CRISPR tech is that it makes animal experimentation faster and cheaper, and will accelerate basic biological research. We still don’t know what the majority of indivdual genes do, let alone how they work in complexes and networks in cells.

STAR-Fusion is a program that detects RNA fusions events in RNA-Seq data. According to the paper describing the program, STAR-Fusion is much better than the dozen or so other callers under active development.

Still, reading the paper left me with some questions. As descried by the authors, STAR-Fusion is not just a good caller, but the best caller by a wide margin. See Fig 3A. The next nine best callers have AUC values of 0.5 to 0.3, but STAR-Fusion has a value of 0.8 in the author’s testing.

And what is the source of this incredible result? The authors are silent on the subject. They don’t know, or perhaps didn’t notice how remarkable their achievement is, and so don’t remark on it. The description of the STAR-Fusion algorithm seems very similar to the algorithms used by every other RNA fusion caller. Some do better than others, so details of implementation must matter.

So what is the critical advance STAR-Fusion makes? Is better sequence alignment key? Is the filtering approach? The paralog handling seems like it cuts down on false positives, is this key? Discovering the critical factors for RNA fusion calling would be an important result.

Or are the performance results in the paper dependent on the synthetic test data set the authors use? Will subsequent papers comparing STAR-Fusion to other methods find that it is only average, or sub-par?

Picked up a GeneTac 1000, a biochip scanner. Here are teardown pictures:
Here is the unit:

Teardown pictures. The unit has a self-contained lamp module that plugs into the main controller unit (EG&G Optoelectronics, Model # 300mXT-04, lamp module LM-300MX). Can’t find much about it–looks to be a 300W lamp. The center of the unit has a CCD camera, a Nikon lens, and a big custom lens, and two sets of filters. The slide carousel is on the other side.

I picked up two ABI PRISM 377 DNA sequencers. These are the last generation of slab gel sequencers.

With the front open, you can see the place where the gel gets mounted.

The left side opens, and the bottom cover comes off. The laser can be seen at the bottom, and some of the power supplies on the left.

Here is the laser, a Uniphase Argon laser, 0.5W 2214-40MLA 1998.

The power supply modules are located on the left side. On the top left is the laser power supply. The electrophoresis power supply is on the left in the middle. To the right of it is the power distribution center–plugs for the laser and electrophoresis power supplies, and the blower motor. On the bottom at the left is the blower motor. In the middle at the bottom the top of a mirror module that bounces the laser back to the right at the level of the bottom of the gel. On the right at the bottom, the servo motor that moves the detector unit along the bottom of the gel.

Here’s the laser power supply. The laser is not plugged in.

Close up of the laser power supply, Uniphase 2114B-40MLA 12A:

The power distribution center labels: J41 DC Power Supply Max 4000W J40 Heater and Pump Control

The electrophoresis power supply: Spellman P/N X2094 Rev. E4 Model No PTV5P300X2094
230V 5A, output 0-5kV, 0-60mA.

Here is a closeup of the servo: Telcomm brushless servo motor

On the back side of the machine behind the top panel is this circuit board, the control, data processing, and interface board.

Two interesting chips on the board, a FPGA and a pair of voltage converters.

The FPGA is a Xilinx XC3064A

The voltage converter.

On the left of the main board is region with cooling lines:

Ion torrent’s sequencer will be coming out this year. The specs:

$50k for the machine.

Runs will generate 100,000 reads of 100 bps, 10Mb per run.

Runs take about an hour, and cost $250 for the chip and another $250 for the other consumables.