Monthly Archives: October 2014

NEJM on Ebola

The New England Journal of Medicine has an Editorial criticizing the quarantines in NJ and other states.

Health care professionals treating patients with this illness have learned that transmission arises from contact with bodily fluids of a person who is symptomatic — that is, has a fever, vomiting, diarrhea, and malaise. We have very strong reason to believe that transmission occurs when the viral load in bodily fluids is high, on the order of millions of virions per microliter. This recognition has led to the dictum that an asymptomatic person is not contagious; field experience in West Africa has shown that conclusion to be valid. Therefore, an asymptomatic health care worker returning from treating patients with Ebola, even if he or she were infected, would not be contagious.

In the same issue, there is an article: Clinical Illness and Outcomes in Patients with Ebola in Sierra Leone. Take a look at the supplementary material. Figure S8 shows temperature and heart rate for fatal and non fatal cases

Panels B, C,E and F represent cases with a normal temperature-pulse association.

All of these are infected. One of the three was a fatal. Table S5 shows symptoms. 11/36 did not present fever. We don’t see if they presented other symptoms but, from the legend:

Eight fatal subjects and one nonfatal subject showed no reported symptoms on the case notification form and were excluded from these results.


From Schieffelin et al. (2014) Clinical Illness and Outcomes in Patients with Ebola in Sierra Leone NEJM DOI: 10.1056/NEJMoa1411680 Figure S7

Perhaps people can die of Ebola without being viremic to the level needed for infect others? I also wonder if they really meant “millions per microliter” or “millions per milliliter”. The former is 109/milliliter. Figure S7 shows viral loads in fatal and nonfatal patients, and it does look like Ebola in serum can reach that level, but two of the fatal cases are well below that. Other sources have claimed that the number of particles needed for an infection is on the order of 1-10. Even at 106/ml,a microliter would be 1000 particles. When people are exposed to infected bodily fluids, are the volumes involved in the picoliter range?

Update: the ID50 is on the order of 1-10 for animal models, but in mice the sensitivity depends on the route of introduction.

The LD50 of mouse-adapted EBO-Z virus inoculated into the peritoneal cavity was ~1 virion. Mice were resistant to large doses of the same virus inoculated subcutaneously, intradermally, or intramuscularly.

Your spambot needs debugging

Found an amusing spam in my comment moderation queue. Excerpt:

{I have|I’ve} been {surfing|browsing} online more than {three|3|2|4} hours today, yet I
never found any interesting article like yours.

{It’s|It is} pretty worth enough for me. {In my opinion|Personally|In my view}, if all
{webmasters|site owners|website owners|web owners} and
bloggers made good content as you did, the {internet|net|web} will be {much more|a lot more} useful than ever before.|
I {couldn’t|could not} {resist|refrain from} commenting.
{Very well|Perfectly|Well|Exceptionally well} written!|
{I will|I’ll} {right away|immediately} {take hold of|grab|clutch|grasp|seize|snatch}
your {rss|rss feed} as I {can not|can’t} {in finding|find|to find}
your {email|e-mail} subscription {link|hyperlink} or {newsletter|e-newsletter} service.
Do {you have|you’ve} any? {Please|Kindly} {allow|permit|let}
me {realize|recognize|understand|recognise|know} {so
that|in order that} I {may just|may|could} subscribe.

College Football Week 6: Chaos

My teams went 0-3 in a series of frustrating losses, each of which was depressing in a different way. But these were just a small slice of a crazy weekend in sports.

Miss State d A&M

The question going into this game was whether the Arkansas game showed A&M’s grit and ability, or revealed problems. Meanwhile Miss State was coming off a bye week after their historic win at LSU. Pundits were picking the Bulldogs despite A&M’s #6 ranking, and the line moved from the Ags being slight favorites to slight underdogs.

The game was in the early slot, starting at 11AM CDT. MSU was missing starting center Stomping Dillon Day, but at gametime it was announced that they would not play one of their better receivers. The Ags revealed that the shoulder injury that Malcome Kennedy sustained against Arkansas would keep him out of the game in Starkvegas. Aggie fans were surprised that among the large numbers of receivers on the roster, walk on Boone Niederhofer was the top choice to take Kennedy’s spot.

The Ags started fast again, scoring a TD on their first drive. State answered with a TD of their own, but with their drive including a Myles Garrett sack of Dak Prescott and the need for some penalties to keep the drive alive, I was hopeful that the Ags would come away with the win.  But after starting fast again, the Ags repeated their problems of keeping drives going. A muffed punt after the second A&M possession gave the Ags an opportunity to retake the lead despite stalling on the second drive, but could not convert on 4th and 1 after a horrible spot on the 3rd down catch by Niederhofer. The D got a stop on the next MSU possession, but the Ags stalled again. The number of drops was something between 9-15 depending on who was counting. Meanwhile, although the D was able to get enough stops to win if A&M was the scoring machine we expected, MSU and Prescott got their act together, found weaknesses, and made it 21-7.

It got worse when Hill threw picks on the next two possessions, squandering a forced turnover after the first interception. This made the halftime score 28-10 after the Ags added a late FG. The D got stops on the first two possessions of the second half. But the O couldn’t shift the momentum by scoring, and when MSU scored on their third possession to make it 34-10, the chances for a comeback faded.  A great catch by Speedy Noil cut it to 34-17, but we couldn’t stop State and another TD made it 41-17. The Ags scored the only points of the 4th quarter to make the final score look better, but it was still a beatdown.

As I write this, Gabe Bock and Billy Liucci are dissecting the game on the radio. The question they are addressing is whether the WR corps got surprised the last two weeks about the difference in play in the SEC West compared to the previous opponents.  This sounds right to me… dropped passes aren’t just about what a WR does with his hands at the end of the play. Everything that goes into running a good rout and creating a window affects the probability of a catch. I once read that running full speed decreases visual acuity due to the amount of head movement that happens in a full out sprint. This 2009 WSJ article about Larry Fitzgerald has some interesting thoughts about what makes a great WR in terms of vision.

Northwestern d Wisconsin

Wisconsin and Stanford both had 2:30 CDT kickoffs, and between being depressed by the Ags taking the dog out, and flipping to other games, I didn’t watch all of this game. The Badgers have had trouble with their passing game all year, which puts extra pressure on the run game. Despite the poor passing, Melvin Gordon has had an exceptional year, and Wisconsin has managed to win against weaker opponents.

If A&M fans are concerned about QB play, they only need to watch the Badgers to see how good we have it. On Saturday the Wisconsin coaches decided that a below average Joel Stave is a better choice than a horrendous Tanner McEvoy at QB. Unfortunately, this improvement includes the fact that Stave’s bad passes are close enough to the receivers to be picked off, while McEvoy tends to miss so badly that even the DBs can’t find the ball.

I missed most of Melvin Gordon’s runs on the way to a career-high 259 yards. Unfortunately for the Badgers, Northwestern’s D kept Gordon from making it into the end zone more than once. The Wildcats took a 10-0 lead into halftime. Gordon’s only TD of the game cut the lead to 10-7 and the hope was that the Badgers would get another game where the run game wears the defense out for a late win.  And when the Badgers stopped Northwestern at midfield after Gordon’s TD, it looked like that would be the script for the day.

Instead, starting deep in their own territory after a penalty on the punt return, the Badgers ran Gordon twice to make it 3rd and 5 from the 11. Instead of taking their chances with another run, the Badgers threw for the sticks… and Stave was picked off at the 16. One handoff later it was 17-7 Northwestern.

A missed Wisconsin FG and a good NW FG made it 20-7. The Badgers got down to the NW 16, but penalties and losses gave them 4th and 21 at the 27. Instead of trying another FG, the Badgers played for field position by taking a delay and punting. This looked like it had paid off when the D held and Wisconsin got the ball back at the Wildcat 44.

Most of the above is reconstructed from the online play-by-play. I only saw glimpses as I channel surfed. I did, however, watch live as the Badgers lined up with first and goal from the 3 with a chance to cut the lead to 6.  What happened next was one of the worst combinations of play calling and execution of the weekend. With everyone watching anywhere expecting the Badgers to blast into the end zone with 3 chances to make 3 yards, OC Andy Ludwig calls a pass. Stave gets chased out of the pocket and goes to the right. Instead of throwing the ball away to take 2 chances to blast Gordon or Clement into the end zone, he tries to hit a receiver for the TD and throws a pick.

Later in the 4th, Stave hit Kenzell Doe on a long TD to get the lead to 6. But now there was only 4:16 to go. The Badgers kicked deep but weren’t able to prevent Northwestern from draining the clock to 33 seconds. A final interception gave Stave 3 for the day, to go with 1 by McEvoy. Wisconsin QBs finished the day with QBRs of 10.1 and 8.8.

Notre Dame d Stanford

While the Wisconsin debacle was going on, Stanford was in South Bend playing Notre Dame in the rain. While it could be viewed as a good old-fashioned defensive game, I’m not sure if the lack of offensive production was good D or below average offense. The Irish outgained the Cardinal 370 yards to 205. With those numbers, it’s not surprising that neither team had a 100 yard rusher. Both teams tried the run 32 times (including sacks); the leading runners where Wright for Stanford with 29 yards and McDaniel for the Irish with 41.

After a 7-7 halftime tie, both teams were futile in the third quarter. Halfway through the final period, ND broke the tie with a FG. Failure to have the holder wear gloves on an earlier attempt in the pouring rain kept them from taking the lead earlier. Stanford put together what should have been the winning drive starting with a nice Ty Mongomery kickoff return to the Stanford 42. A mix of passes and runs got the Cardinal to the ND 33, where QB Kevin Hogan converted a 3rd down by lofting a 23 yard pass to Devon Cajuste at the 10. I was kind of shocked that this was completed. The ball floated high into double coverage and the ND defenders just failed to break up the completion. After two futile running plays made it 3rd and goal from the 11, Notre Dame dropped into a pass-prevent defense, leaving a vast hole up the middle that Remound Write ran through pretty much untouched for a TD and the lead with 3 minutes to go.

With Stanford’s top ranked defense and the offensive futility displayed through the day, that should have been enough. But instead the Cardinal put in a textbook display of how to blow a late lead.  They started by booting the kickoff out of bounds, giving the Irish the ball on their own 35. They then showed why people say that prevent defense is only good at preventing victories. Using a 3 man rush, they gave ND QB Everett Golson time to find holes in the secondary and march down the field. Worse, if you’re going to drop 8, it helps if they pass defenders actually know where to go. One busted coverage led to a pass interference call, giving ND the ball on the Stanford 22. The second busted coverage came on a play that was a golden opportunity for the D to save the game: 4th and 11 from the 23. Instead, the D left a tight end wide open in the left side of the end zone. TD Irish.

Stanford got the ball back with a minute to go and managed to get to midfield. But Hogan took an intentional grounding penalty and the runoff ended the game with a whimper. From the post-game.

Stanford coach David Shaw was asked what coverage the No. 14 Cardinal were in on the play.

“There was no coverage on Notre Dame’s touchdown pass,” he said. “That sounds sarcastic but he was wide-open. There was nobody on him.”



A lot of teams thought they had an opportunity to move up in the rankings when unranked Arizona upset #2 Oregon on Thursday night. Not so much.


Sous vide tandoori-style lamb chops

Debby got some leftover vegetable biryani from a Biology party.  We’re going to use as a side with some lamb chops I got from Rosenthal, so I thought I’d do something different and make them with a tandoori marinade.


  • Yogurt
  • Garlic
  • Ginger
  • Cumin
  • Coriander
  • Cinnamon
  • Garam Masala
  • Sambal
  • Some of the goop from a jar of Indian mixed pickle
  • Salt

Mixed in a stainless bowl and put thawed lamb chops in at about 2:40 PM. The color is sort of a light chocolate brown.

Sous vide and broil

Let it warm to room temp and bagged it all. 135F for 2 hours starting at ~6PM. Finish under the broiler.

It might have been better to finish in a pan or on a grill, as the broiler may have taken them past medium rare. but the marinade worked well.

College Football Week 5:Comebacks

I didn’t get around to posting last week’s results, when my teams went 3-0, but each of the wins foreshadowed problems that surfaced this week. I meant to write something last week, but got busy with actual work.

Wisconsin d S. Florida

The Badgers played an early (11AM CDT) game against S. Florida. The game was tied at 3 at halftime, as Wisconsin continued to show problems in the passing game, and was slow to get Melvin Gordon and the passing game going with Tanner McEvoy. The Badgers broke out in the third quarter. Final Score: Wisconsin 27, S. Florida 10.

A&M d Arkansas

The Aggies went to Dallas for the second straight week to face an improving Arkansas team in their SEC opener (talking heads seem to forget that we had already played S. Carolina). After the usual fast start, A&M fell behind on a long run by RB Alex Collins. A&M tied it on a TD to Ed Pope, but the Ags misfired with dropped passes and a surprising FG miss by Lambo. While the Ags drives misfired, we did an OK job stopping the Razorbacks until a fake punt that went 51 yards to give Arkansas a 21-14 lead going into halftime. With Arkansas getting the ball first to start the second half, A&M fans were worried about getting into a shootout needing to break service. The D held on that first drive and A&M moved to the Arkansas 21 with a chance to tie the game. But a Kenny Hill to Ricky Seals-Jones out pattern on 4th and 1 was well covered, and the Ags gave the ball back. Arkansas got a 2 TD lead on their next drive on a play-action pass to TE AJ Derby. Derby blocked CB Devante Harris before releasing into the open with no safety help, while QB Brandon Allen did an excellent job of hiding the ball after faking a handoff. 28-14 Arkansas.

The first major turning point of the game came on a bonehead play by Arkansas tackle Dan Skipper in the 4th quarter. Arkansas ran a well-blocked play that sprang Jonathan Williams to the 2 yd line. But away from the point of attack, Skipper was knocked down and reacted by not one but at least two flails of his legs, ultimately drawing a tripping call that brought back the big play that would have led to a 3 TD lead. A&M’s next possession after a punt to the 12 yard line took 2 plays. The second was an 86 yard bomb to Pope to cut the lead to 28-21. Arkansas went 3 and out on their next drive, but Kenny Hill threw a long interception to give the Hogs the ball back with 8:34 to go. Both defenses got stops and Arkansas took over with a TD lead with 5:27 left. A pair of 17 yard runs got the ball into A&M territory. A fumbled snap was recovered by Arkansas, but the Hogs settled for a FG attempt from the A&M 31 to make it a 2 score lead with less than 3 minutes to go. Unfortunately for the Arkansans, the FG was missed. The Ags struck quickly again, with Hill hitting Josh Reynolds on the right side between two defenders. Reynolds shot through the gap and ran to the end zone to tie it up.

After stopping Arkansas again, the Ags had a chance to win in regulation, getting the ball on the 25 with 1:18 left. A&M got as far as their own 38 before opting to go to OT. The OT didn’t last long, as the Ags scored in one play on a Hill pass over the middle to Malcome Kennedy (who got away with a false start that was disguised by Hill’s acting like there was an audible after Kennedy’s twitch). The Arkansas linebacker bit on a fake WR screen to the right, and Kennedy ran straight up the middle to snag the pass and run into the endzone. Arkansas had to score a TD to keep OT going, but couldn’t convert on 4th and 1. Final Score: A&M 35 Arkansas 28. Bullet dodged.

Stanford d Washington

The Cardinal went to Seattle to play Washington. The game started while the A&M game was already going, so I only caught glimpses of it. But what I did see seemed like typical Stanford for this year: good D combined with an offense that has trouble in the red zone. Stanford had 3 turnovers, and the game was tied at 13 in the 4th quarter. I did not see the peculiar decision by Washington to run a fake punt on 4th and 9 from near midfield. It was stuffed and Stanford scored the game winning TD on the next possession.


There were lots of interesting results on the field, but the story that dominated the week was a decision by Michigan’s Brady Hoke to sub in a player who was concussed during a blowout loss to Minnesota. This seemed to be a product of cluelessness and incompetence more than a manifestation of macho throwback values, but it was still appalling. Michigan handled the aftermath badly as well, leading to campus rallies against the AD and grandstanding from congresscritters.

On the field:

Sous vide halibut

Debby bought some frozen halibut filets. Thought I’d try to cook them sous vide.

sourcetemp and timeNotes
Sous vide Supreme132F 20 min
beyond salmon128F 15 min/inch
Chefsteps forum113–122 °F 30 min or so
Codlo guide109F rare
122 medium rare
140 medium


Decided on 128F. Salted the frozen filets and put them in a bag with some butter and lemon slices. Into the bath at 7:45PM. I figure to add some time to the recommended 20-30 min for the fish to thaw.

The bag didn’t sink as much as when I do red meat, so I put a spatula in to make sure the bag was submerged.

As the Giants and Nationals went into extra innings in game 2, I took the fish out of the bag and portioned it. I had intended to pan-fry the portions to crisp up the skin, but it was too fragile. Instead, I removed the skin and served with rice and mixed vegetables.

The fish was tender and moist – perhaps to the point of being on the wet side. It was tasty but a bit on the boring side.  Despite the lemon slices and the spritz of lemon juice I added when I served them, I think this would be better with some sort of bright and spicy sauce.

But for a meal put together from what we had in the fridge and freezer on a night when we needed to go grocery shopping… not too bad.

Dealing with ShellShock on our older mac servers

We have a couple of machines that are still running Snow Leopard, so the Apple patch won’t work. One option is to recompile bash and its patches from source, but why do that when I already have MacPorts on those machines.

Testing the vulnerabilities

From Stack Exchange, there are multiple vulnerabilities:
The first one is tested with this:
$env x='() { :;}; echo vulnerable' bash -c 'echo hello'
if the shell is vulnerable, it will echo vulnerable and hello. Otherwise it just echoes hello.

A second vulnerability is tested with
$env X='() { (a)=>\' sh -c "echo date"; cat echo
if you’re OK, you will see something like
cat: echo: No such file or directory

sh: X: line 0: syntax error near unexpected token `='
sh: X: line 0: `X () { (a)=>\'
sh: error importing function definition for `X'
cat: echo: No such file or directory

The important thing is that you don’t want to see an actual date and a file created called echo.

The third vulnerability redefines the ls command
$ env ls="() { echo 'Game over'; }" bash -c ls
Vulnerable systems will echo Game over.

Installing the fix

Getting it installed is easy after updating to the latest versions of the ports
sudo port selfupdate
sudo port upgrade outdated
sudo port install bash

But this only applies the MacPorts bash for the user. To make it the default, we have to make it the default shell for an intruder. To do this we need to edit /etc/shells to replace /bin/bash with /opt/local/bin/bash.

I think this is enough. But I still have some concerns. The bash version in MacPorts seems to be 4.3.28. Apple’s official release after the patch is 3.2.53 (apparently this is equivalent to 3.2.54 in terms of patches). The MacPorts version is still vulnerable to the second problem. I’ll be watching MacPorts for updates, but I am also looking at whether the machines, which are too old for Mavericks, can be updated to Lion or Mountain Lion.

Update: MacPorts pushed another bash update over the weekend

Edge wander

In class today, we talked about the first Assemblethon paper. A student asked about the term “edge wander”, which comes from a paper by Ian Holmes and Richard Durbin. Figure 6 from the paper illustrates the basic idea.

Edge wander is a problem in multiple sequence alignments, and often scientists manually adjust alignments based on some heuristics that are not entirely clear to me. At last year’s Texas Protein Folding and Function Meeting, Patsy Babbitt mentioned in passing that manually adjusting multiple sequence alignments has become impractical as the number of available sequences in conserved protein families is exploding.

More ribs

These were from last week. Bought as a rack from Rosenthal. Treated with salt pepper and brown sugar for a day or so. Sous vide at 138 for about a day and a half. Finished under the broiler